User Guide

  Strategies Management Page

• What is a Strategy?
A Strategy, in the context of AlphaDroid Strategies, is a set of up to 12 mutual funds or ETFs along with an indicator algorithm to evaluate the performance of each to determine which one, and only one, is currently demonstrating leadership and should be owned. Click the picture to the right to view the layout and functions of the Strategies Management page. This is where your investment Strategies are managed. To go to the Strategies Management page, click the Account Management tab on the top menu, and then the Strategies tab on the second level menu. The Strategies on your page are yours alone to manage and edit.

A Custom Strategy is simply made by editing any of the ticker symbols in an existing Strategy, adding new ticker symbol to a Strategy, or deleting an existing ticker symbol from a Strategy. All of these actions are initiated by clicking the ticker symbol (or blank) to open the popup window to "Find a Mutual Fund, or ETF." The popup window enables you to search by either ticker symbol or by words in its name. To add it to your Strategy, click the button. The popup window also provides a means to delete the current ticker symbol by clicking . As shown above, when you place the mouse pointer above a ticker symbol, its name, pertinent hold days, and early trade fees are shown. At each step of your Strategy development you can check its progress by clicking the chart icon, and selecting other properties of the Strategy by clicking the information icon. 

Making a Strategy-of-Strategies
A Strategy-of-Strategies uses special ticker symbols to refer to your other Strategies. As with a normal AlphaDroid Strategy, it will pick the one, and only one, best performing Strategy from among up to 12 candidates. Use the ticker symbol format of ADnnn, for AlphaDroid# nnn, where nnn does not have a leading zero if it is less than three digits. You can also mix normal ticker symbols with the Strategy ticker symbols if you like. When you create a Strategy-of-Strategies, there will be a radio button to designate it as such. All Strategy-of-Strategies must reside in Strategy rows #501 and higher, and all normal Strategies must reside in Strategy rows #1 to #500.
 
 
Algorithm Automation Simplicity!
Although indicator algorithms can be complex and difficult to understand and configure, the good news is that we have totally automated the indicator algorithm configuration for each Strategy according to the character of your chosen funds. There are significant differences in the behavior of mutual funds, bonds, and stocks. For example, commodities, such as oil and gold, will behave differently from the stocks of companies that manufacture consumer goods, or from funds holding T-bills or certificates of deposit (money market funds). AlphaDroid automatically determines the algorithm and parameters most appropriate for the character of each Strategy so you don't have to. That's why we say "Our computers will do the hard work while you go have a life!" Just throw a set of funds at it, and AlphaDroid automatically figures out how best to treat them.

Strategy Performance: Score- Safety-MDD
These performance metrics are shown with each Strategy for easier comparison. The definition of these performance metrics can be found in the Score, Safety, and MaxDrawdown sections of this User Guide.

Add Strategies to Your List
The Strategies Management page contains your personal list of up to 10 Active Strategies for which Trade Alerts are sent and may require a subscription fee, and up to 20 Sandbox Strategies which are for experimental evaluation, but neither send Trade Alerts nor require a subscription fee. There are three ways to add a Strategy to your list:

1. Select ready-made Strategies by clicking the Strategy icon and considering the Strategies presented in the
    Select-A-Strategy popup window. Click the row of the Strategy to select it, then click the button
    at the bottom of the page to add it to your Strategies list.

2. Import a Strategy using a Strategy-ID by clicking the Strategy icon, pasting the Strategy-ID into the text
    box at the upper right. Then click the   button above the text box. To send a Strategy-ID to a
    friend, click the Information icon to find the Strategy-ID, then copy and paste it into an email. 

3. Assemble your own Custom Strategy by clicking the ticker symbol positions in the Strategy List and selecting
    one from the many thousands of mutual funds or ETFs on the list that pops up. Additional details
    and tips for building great Custom Strategies are detailed below.

View Trade Signal History for a Strategy
You may view the Trade Signal History for a Strategy by clicking the History icon. The Trade Signal History, as pictured to the right, should not be confused with the Actual Trade History found in the Trade Information section of the page. The Trade Signal History is what the Strategy algorithm says one should have done to achieve the results portrayed on the charts.

The list shows the entire history of trade signals, days held, performance for each holding compared to the S&P 500 and the cumulative value for both the Strategy and S&P 500 had you invested $10,000 in each at the start of the available data for the Strategy.

The generic symbol $CASH is used when StormGuard directs you to move to the safety of a money market fund. Not everyone has access to the same money market funds, and they are all pretty much like "peas in a pod", so whatever your brokerage has available will work just fine.

At the bottom of the listing you can click the button to download a CSV (comma separated variables) spreadsheet of all the important Strategy statistics and trade signal history. Add "/d" to the end of your Strategy Name to further add daily data for the Strategy and the reference fund. (Note: make sure the "/d" shows up in the name on the chart because if the name is too long and it is cut off the extra data will not be generated.)

Correct Interpretation of History Data: A common misinterpretation of the Trade Signal History data is that each line reports performance associated with the BUY fund. The correct interpretation is that it reports the status on the date found in the left hand column. On the day of the Trade Signal Date, one has only just bought the BUY fund. Thus the performance data reported relates to the SELL fund. A good reminder of this is the 4th column title "Days Held (fund sold)."  Thus, when you click to view the example above-right, the proper interpretation is: On Jan 31st 2011 AlphaDroid signaled to BUY EWY and to sell IIH which had been held for 248 days. During the 248 day period from May 28, 2010 to Jan 31, 2011, the S&P 500 returned 19.5% and the Strategy returned 23.0%.

View Strategy Performance Charts
You may view the current performance charts for a Strategy by clicking its Chart icon. In a single view, five separate charted measures of risk and return performance are provided, as described in the About AlphaDroid Charts section below. Fresh charts are made within a few seconds whenever you make a change to a Strategy to enable evaluation of the change's effects. Fresh charts are also produced: (a) during daily data update processing, (b) when a Strategy is imported to your list, and (c) when you change a fund in the Strategy or change its Minimum Hold Time (see below).

Strategy Information
The Strategy Information popup window contains these elements:
 • Strategy Name editing so you can rename it to something meaningful to you.
 • Strategy Classification to indicate the investment style. For your reference only, it doesn't affect the Strategy.
 • Strategy Minimum Hold Time to control when a Trade Alert can be issued. (See notes just below.)
 • Strategy-ID  for you to copy and paste into an email so a friend can import an exact copy of the Strategy.
Note: The Information icon will be replaced with the warning icon if AlphaDroid detects a problem with one of your Strategies (such as one of the funds going defunct). The warning message will be displayed when the icon is clicked.

• Minimum Hold Time
There are seven choices for the Strategy Minimum Hold Time, as listed below. You can select one in the Strategy Information popup window by clicking the Information icon to the right of the Strategy name.  This parameter can be used to make AlphaDroid's algorithms comply with minimum hold time requirements, whether imposed by a particular fund, or by a particular investment plan. For example, most mutual funds and ETFs have no minimum hold time. However, Fidelity Sector Funds all have a 30-day minimum hold time with a 0.75% early trading fee, most Fidelity International Funds have a 90-day minimum hold time with a 1.5% early trading fee, and all 529 College Savings Plans have a one-year government mandated minimum hold time.

AlphaDroid maintains a file of the minimum hold time and early trading fee for hundreds of popular funds. However, there are many thousands of funds available. If we have the information for your fund, it will be listed in the right column of the Find-A-Fund popup window on the Strategies Management page, or in the screen tip text that appears when your mouse pointer is placed over the ticker symbol to display its name. When AlphaDroid knows the minimum hold time for a fund, it is incorporated into the trading algorithm only if the "Trade Automatic" setting is selected. Currently we have the minimum hold information for Fidelity, Vanguard, and American funds. In the future, on request, we will add such information for other funds of importance to you to support your custom Strategies.

AlphaDroid generates a new Trade Alert when trend leadership changes according to the Hold Time Option selected:
 • Trade Automatic - Trade alerts issue at month, but will wait additional months if a fund hold time is > 30 days.
 • Trade Any Day - Trade alerts issue immediately, but you'll be warned if a fund hold time is violated.
 • Trade Weekend - Trade alerts issue on the weekend only, but you'll be warned if a fund hold time is violated.
 • Trade Month End - Trade alerts issue at month end only, but you'll be warned if a fund hold time is violated.
 • Trade if > 30 days - Trade alerts issue if current fund has been held at least 30 days.
 • Trade if > 60d - M.E. - Trade alerts issue on Month End if current fund has been held at least 60 days.
 • Trade if > 90d - M.E. - Trade alerts issue on Month End if the fund has been held at least 90 days.
 • Trade if > 1yr - M.E. - Trade alerts issue on Month End if current fund has been held at least 1 year.

Note: Trade Automatic is the default setting for a good reason. Not only does it watch out for special hold time rules for funds, but it also trades month-end, which almost always performs better than Trade Any Day because it can be shown that the trend signals actually do become more reliable around the end of the month when many fund managers start making next month's changes. This is documented in the Hurst Exponent Fingerprint section of the Sector Rotation Theory page. If you are anxious to have a faster responding system, please read about The Faster Response False Dilemma.

Note: Changing Minimum Hold Time will cause the algorithm to re-evaluate and determine what the best parameters are for the Strategy, and could result in using a longer or shorter trend measurement period, and consequently a possible change in the one designated as the trend leader. Determining the trend leader depends heavily on the period of time over which the trend is measured. Imposing a holding period imparts a delay effect much like changing the trend measuring time constants, thus it should be expected that changing the Minimum Hold Time rule will case a change in the optimum trend measuring time constants.

• Fund Ticker Symbols
By clicking the position of a ticker symbol, the Find a Fund or ETF popup window will appear and enable you to search by either ticker symbol or by words in its name. To add it to your Strategy, click the button. The popup window also provides a means to delete the current ticker symbol by clicking . As shown above, when you place the mouse pointer above a ticker symbol, its name, pertinent hold days, and early trade fees are shown.

Mutual Funds & ETFs: There are in excess of 12,000 mutual funds and ETFs available to choose from. Please contact us if there is a mutual fund or ETF that is important to you but is not on the list. We will do our best to make it available provided it has at least 3 years of data. The reason for requiring 3 years of data is that AlphaDroid's algorithm cannot properly characterize a fund that has not seen a variety of market conditions, and you risk the possibility of unexpected Strategy behavior in the future. If you must have a fund or ETF with shorter duration history, you may be in love with it for the wrong reason. The right reason is having increased probability of higher returns and decreased probability of loss. A sexy name and low introductory fee is not a sound reason for a financial marriage — long-term character matters.

Note: Ticker Symbol Changes Causes Re-Optimization: Each time you change one of the ticker symbols, AlphaDroid will re-make your Strategy chart so you can immediately see the effect of your changes. Note: Changing any ticker symbol or the Strategy minimum hold time causes AlphaDroid to re-evaluate and determine what the best parameters are for the Strategy, and could result in using a longer or shorter trend measurement period, and consequently a possible change in the one designated as the trend leader. Determining the trend leader depends heavily on the period of time over which the trend is measured. Performance ranking last week is not the same as performance ranking last month. Consider that there are many routes to your favorite restaurant, and right now there is an optimum set of roads in the route. But that may change if some roads are closed and new ones are opened, or if you impose a rule for how frequently you can make a turn. Likewise in a Strategy, there are often multiple sequences of ticker symbols that produce fairly similar results, and small changes to the Strategy can cause a different sequence to become the optimum sequence.

Extended History Ticker Symbols The data history length of each of your Strategy ticker symbols can have a profound effect on the projected performance of your Strategy. For example, if you are modeling a Strategy with the original 9 SPDR sector ETFs and you want to see if additional safety is afforded by adding either IEF or TLT (treasury fund ETFs) to the Strategy, you will quickly note that neither of these ETFs has data that extends back past mid-2002 and thus they cannot suitably model performance of the intended Strategy during the 2001-2002 market crash. Matters are even worse if you want to use the UBT (2x treasury ETF) because its data does not start until January 2010, thus making it impossible to model performance during either of the two recent market crashes.

One solution to the problem of satisfactory modeling is to create artificially extended versions of important modeling symbols.  Some of these symbols can be artificially extended by adding on older data portions taken from another fund that can act as a proxy for this fund during earlier periods. Leveraged 2x and 3x ETFs require the use of a scale factor when extending them using 1x data.  However, 2x and 3x ETFs have other characteristics that require a scale factor somewhat less than would be expected from its name. The main cause of this problem is volatility, which causes daily rebalanced ETFs to decay over time (go ahead ... Google that). The scale factors we used were derived from actual performance comparisons during generally rising markets (which is the only time the relative scale factor matters with True Sector Rotation). These extended ticker symbols have a "-" added to them to indicate they are extended versions. A list of the available extended history ticker symbols follows:
 

Extended History Ticker Symbols 

Tip #1. One of the better uses for the  long-term treasury ETFs is with StormGuard as the Bear Market Symbol so that when StormGuard triggers, instead of going to the safety of $CASH (your favorite money market fund), you will instead receive a trade alert to buy the specified Bear Market Symbol. Since long-term treasuries are negatively correlated to the S&P 500, you can generally expect a bit of a performance improvement when implemented. Click to expand the Treasury ETFs comparison chart (right) showing the significant difference between these ETFs, and their differences in character from the S&P 500. Numerous ready-made Bear Market Strategies have been designed to span the range of aggressiveness and address the problem that long-term treasuries have not always been negatively correlated to stocks.

Tip #2. The 1x ETFs in the pink section were extended to provide better Strategy modeling in conjunction with Strategies using SPY, MDY, and the original SPDR sector ETFs, all of which have histories that go back to the mid/late 1990s.

Tip #3. The 2x leveraged ETFs in the grey section and the 3x leveraged ETFs in the orange section were extended to have a common start date of 1/2/2004 so they all could participate in the initial tuning of the Strategy, as opposed to effectively tuning for one set of ETFs, and then starting forward walk with a different expanded set of ETFs that may not play together well in the same way. Your objective, as a Strategy designer, is to provide the algorithm with a truly representative sample of Strategy characteristics prior to the specified BornOn Date so that it can properly tune itself.

Tip #4. The VIX futures contract ETNs in the bottom section aren't for the faint of heart. We've posted an example Strategy on the popup list entitled "A Nose for VIX." While it has posted strong returns, its volatility is quite high. It may inspire contemplation (over a glass of red) whether these funds have completely crossed the line from that of investing to gambling. There is not even a remote connection to owning a piece of a company's assets or earnings. Ah, but in the end, is it really any different from owning a pack of faceless companies in a fund that you trade for another pack in just 30 days proving you really are just betting on the froth rather than loyally supporting development of a worthy, innovative corporation?

• BUY/SELL Trade
This is where the trade details can be viewed, including: (1) the trade signal date and BUY/SELL ticker symbol information, (2) a red button to acknowledge completion of the recommended trade and to inform AlphaDroid that you have completed the task so that no further reminder emails will be sent, and (3) a green button to indicate that you have already acknowledged the most recent trade. If you find you acknowledged a trade in error, click the Strategy Trade History icon where you will find the button to undo the trade acknowledgement status in our system (not the actual trade at your brokerage).

$CASH, $WAIT, -NEW-
When StormGuard indicates it is time to move to/from the safety of a money market fund, it uses the generic symbol $CASH for the Buy/Sell ticker symbol because not everyone has access to the same money market fund. When you see $CASH, substitute your favorite money market fund. When creating a new Strategy it is important to choose a good entry point into the Strategy's recommended fund. If there are only a few days until the next likely Trade Alert or if the last recommended Buy ticker symbol has a short-term trend that has gone negative then AlphaDroid's algorithm will use the $WAIT symbol to indicate you should wait a bit for a better entry point into the Strategy. In both case there will be important and specific trade advice offered by clicking the icon. The -NEW- ticker symbol will appear as the SELL symbol for new Strategies that have no prior recorded history of trade acknowledgements to specify what you actually do own and should sell.

Warning or Special Trade Advice
If there is a special situation regarding the recommended trade, the Warning icon appears instead of the icon. When you click the icon, the warning message will be displayed. The message could be generated for numerous reasons, including: (1) the current fund you own may have a stronger trend signal than any of the funds in the new Strategy you have just selected, (2) one of the funds in your Strategy is defunct and should be deleted or replaced, or (3) you have just created a new Strategy and special advice is offered.

Trade Options Information
If there are no special considerations associated with the suggested trade, the Trade Options Information icon appears to the right of the button, to let you know that there are multiple options for how one might respond to the recommended trade. When the Option Information icon is clicked, the following text is displayed:

You have four options for acknowledging this trade: Make the trade and then return here and click the Acknowledge Trade button so AlphaDroid will no longer send you email reminder Trade Alerts.

2. Delay making the trade until a later or better time and don't click the Acknowledge Trade button until then.

3. Reject this Strategy in favor of another Strategy by clicking the Strategy icon and selecting a new Strategy from the list.

4. Exit for the sidelines to sit in cash (any money market fund) for a while by changing to the Free Strategy called
Sidelines In Cash. Click the Strategy icon select the Sidelines In Cash Strategy. 

View Strategy Trade History
You may view the history of trades you have made in a Strategy by clicking the History icon in the Trade Information column. Unlike the "Trade Signal History" controlled by the Strategy algorithm, this is a list of all trades you have acknowledged, rejected, or that still may be pending. It is a log of your activity for this particular "bucket of money," not a list of "trade signals" for the Strategy. If you acknowledged a trade in error, click the button on this screen to undo the trade acknowledgement status in our system (not the actual trade at your brokerage).

Delete a Strategy from Your List
To delete a Strategy, click on the Delete icon on the far right of the Strategy line. You will be asked to confirm that you really do want to delete the Strategy before it is removed. The first Active Strategy has a grey icon indicating you cannot delete the first Strategy - but you can edit it or import another Strategy over it.

  Strategy Charts

• Logarithmic Price Chart
A logarithmic price chart has the advantage of vertical interval spacing that provides the same percentage change for each interval. On a logarithmic price chart, performance is measured relative to its price at the start of the chart — hence everything starts at 0.0% return on the left side and goes from there.

On the chart to the right, the first vertical interval above 0.0% is 41.4%. A 41.4% gain at the first vertical interval grid line means our total account value at this level is 1.414 times as large as the starting value. We have the original unit amount (1.0) plus the return (.414). Thus, if we started with $1,000, we would now have $1,414.

Likewise, the next interval up will be another 41.4% return, compounded on top: $1,000 x 1.414 x 1.414 = $2,000, which is a 100% return on the original amount. With another 41.4% return we would have $2,000 x 1.414 = $2,828, which is $1,828 more than what we started with and a total return of 183% — and so on.

One nice feature of this is that a straight line represents a constant percent return per year. For example, if you drew a straight line along the crests of the bumps of the yellow curve, it would have a slope that takes approximately 11 years to raise 7 intervals (.64 intervals per year), or roughly  .64 x 41.4% = 26.5% per year.

As can be appreciated when viewing this chart, the relative performance of the Strategy's performance (plotted in yellow) can be compared to the S&P 500 index (plotted in white) and each of the constituent funds. By clicking one of the blue buttons in the lower right, the logarithmic chart can be viewed in three different time scales — 3-Years, 10-Years, and Max-Years (where Max is the full number of years in the fund database for at least two of the funds). If fewer than 10 years of data is available, the 10-Year button will disappear.

• Annualized Return Bar Chart
Below the logarithmic price chart is the annualized return bar chart that compares the annualized return (average yearly return) performance of the Strategy to the S&P 500 Index over the most recent 3-Years, 10-Years, and Max-Years. The numerical value for AlphaDroid's performance for each of those periods is located below its corresponding vertical bar.

• Sharpe Ratio Risk Measure
The Sharpe Ratio is commonly used in the financial industry to measure an investment's added return over that of a very safe money market fund relative to the higher investment risk taken. It is named for William Sharpe, Professor of Finance, Emeritus, at Stanford University's Graduate School of Business and the winner of the 1990 Nobel Memorial Prize in Economic Sciences. The mathematical expression for the Sharpe Ratio is:

            Sharpe Ratio = (Fund Average Return - Money Market Return) / (Fund Standard Deviation)

The Fund Average Return is calculated by first finding the ratio of the ending value to the starting value of the fund, finding its Nth root, and subtracting one — where N is the number of years between the starting value and the ending value. The average return for Fidelity SPRXX over the same period (about 4.5%) is used for the Money Market Return value.

               Fund Average Return = (EndValue/StartValue)^(1/N) - 1

The Fund Standard Deviation is calculated by stepping through the database day-by-day and calculating gain/loss from one year earlier less the Average Annual Return, squaring it and summing it to form a TotalVariationSquared value. The TotalVariationSquared is then divided by the total number of days. By taking the square root of this value we get the Fund Standard Deviation. 

• Sortino Ratio Risk Measure
The Sortino Ratio is commonly used in the financial industry to measure an investment's added return over that of a very safe money market fund relative to the higher investment risk taken. The ratio is named for Dr. Frank A. Sortino, an early popularizer of downside risk optimization. The mathematical expression for the Sortino Ratio is:

            Sortino Ratio = (Fund Average Return - Money Market Return) / (Downside Deviation)

The Fund Average Return is calculated by first finding the ratio of the ending value to the starting value of the fund, finding its Nth root, and subtracting one — where N is the number of years between the starting value and the ending value. The  Fidelity SPRXX money market fund is used as the riskless reference return to beat.

               Fund Average Return = (EndValue/StartValue)^(1/N) - 1

The Downside Deviation is calculated by stepping through the database day-by-day and calculating gain/loss from one month earlier relative to the monthly return of the money market fund. If, and only if, the monthly return is less than the money market return is the difference in return (StrategyMonthlyReturn - M.M.MonthlyReturn) squared and added to a TotalVariationSquared value. The TotalVariationSquared is then divided by the total number of days. By taking the square root of this value we get the Downside Deviation. 

• Prudent, Suitable, and Relative Risk
As detailed in our white paper "Satisfying the Prudent Man," strikingly absent from all regulatory documents pertaining to financial advisors is (a) any practical definition of risk and how it is quantitatively measured; (b) any guidance for determining how much diversification is required; and (c) any specification of the risk categories (conservative, moderate, and aggressive) financial professionals most commonly discuss. Fortunately, the traditional risk-classification model portfolios promoted by respected industry leaders can be used to form a de facto consensus set of risk-ranked portfolios (right) that can be modeled, quantified, and used as reference standards in assessing the relative risk performance of other portfolios.

In field of Behavioral Finance, the important contribution by Daniel Kahneman and Amos Tversky in their seminal paper "Prospect Theory" showed that for decisions involving risk investors feel losses hurt more than gains feel good (loss aversion). This lead to the development of Post-Modern Portfolio Theory and the Sortino Ratio, both of which use downside deviation to measure of risk, as opposed to using standard deviation.

Likewise, AlphaDroid's Relative Risk (R.Risk) is calculated as the ratio of downside deviations between a portfolio of interest and the consensus aggressive portfolio. The R.Risk value is displayed on all AlphaDroid charts in the statistics section (right) as well as on the Strategies Management page. A portfolio with an 71% R.Risk would thus have a downside deviation that is only 71% as large as the consensus aggressive portfolio. The downside deviation is calculated using quarterly returns measured daily across the entire time span of the portfolio's equity curve. To ensure consistency of the ratio, the same time span is used for computing the downside deviation value of the consensus aggressive portfolio. Downside deviation is calculated as the standard deviation of all negative returns for all 90 day intervals over the full time span.

 • Probability of Loss of Real Money
At AlphaDroid Strategies, we believe that the two main measures of risk used in the financial industry, the Sharpe Ratio (explained above) and the coefficient of variation (defined as the standard deviation divided by its average value) both miss-characterize risk by ignoring how humans perceive risk. Both use focus on a statistical measure called standard deviation (basically the wiggliness of the line), which is affected by both upside wiggles and downside wiggles. Given that investors pretty much like upside wiggles, it is difficult to make a case for including them in a measure of risk. What scares investors is the possibility of losing real money — money at risk is money that could be lost.

To be meaningful, this measure of risk must include magnitude of loss and probability that it might occur. We determined that a magnitude of about 15% loss is where investors start becoming seriously concerned. As a simplistic example, if a fund experienced a loss of 7.5% year-to-year on average once every 4 years, then our measure of risk should be (7.5%/15%)x(1/4) = 12.5%. The numbers in the upper right corner show the values for the Strategy (yellow marker), and for the reference fund (white marker).

The Chance of a 15% Year-Over-Year Loss is calculated by first stepping through the database day-by-day and calculating the return for each from one year earlier.  If the return is positive then the day is skipped. If the return is negative then it is divided by 15% to normalize/scale it properly, and then it is summed with the losses from other days to form a TotalLoss value.  When all of the days have been checked, the TotalLoss is divided by the number of days that were checked to result in the value for Chance of 15% Loss in a Year.  The Average Annual Return is calculated by first finding the ratio of the ending value to the starting value of the fund, finding its Nth root, and subtracting one - where N is the number of years between the starting value and the ending value. Thus the  Average Annual Return  =  (EndValue/StartValue)^(1/N) - 1  Note that the calculation is performed over the period of time for which data exists for the Strategy ticker symbols and may result in different values for the reference fund in different Strategies. The date range is specified in the title of the risk section on the chart.

• Relative Trend Strength
The Trend Strength is the final singular figure of merit that AlphaDroid generates and uses to determine which one, and only one, of the funds has taken leadership and should be owned. The green-bar trend chart is calibrated in percent return per month and is an indicator of possible returns next month — to the degree that the current trend continues. Although "trend" means that something in the recent past tells us something about the near future, the future is also buffeted by the random events of the world.

Trade Alert vs. Top Trend: It is important to understand that just because a fund/stock makes it to the top of the Trend Chart does not mean that you should instantly run out and buy it. Each Strategy also has a "Minimum Hold Time" rule, such as "Trade Month-End," which determines when the Strategy will actually employ the Trend information and possibly generate an email Trade Alert. In the case of Trade Month-End, only after the market close on the last trading day of each month will the algorithm check the Trend for each of the funds to determine if there is a new leader, and if so, update the Sell/Buy information for the Strategy and send an email Trade Alert. It's not uncommon for one fund to be in the lead mid-month, but another to take the lead near the end of the month before the actual decision is made. Only with the "Trade Any Day" setting will a new email Trade Alert be sent on the exact day there is a new trend leader. Reasons for selecting more restrictive settings include early trade fees for some mutual funds, and that generally, month-end trading actually does perform better as described here. You can edit the Strategy's Minimum Hold Time parameter by clicking the Information icon to the right of the Strategy name on the Strategies Management page.  

Inconsistent Trend Position: It's not uncommon to see a pair of funds in one Strategy have trend ranks become reversed when they are both also in a second Strategy. The reason this is possible (and rational) is that there is no perfect fixed definition for "Trend" that is optimum for everything. One fund may have a better one-week trend, but the other may have a better one-month trend. Thus, positions can swap depending on exactly how the trends are measured. Each Strategy has a custom set of trend algorithm parameters that are determined specifically for the set of funds in the Strategy. If one or more of the funds change, it is likely there will be at least a small variation in what constitutes the optimum measure of trend for the Strategy.

Strategy Editing: Further to the point above, if you change something in a Strategy and then reverse the change, there is no guarantee that the original set of trend parameters will be used again. Parameters are determined in full view of all past market data. Thus, parameters set with data that is current, versus data from one year ago, will likely be a little different. A difference in how trend is measure can result in a slightly different path of fund ownership. Similarly, a change in the Minimum Hold Time will also generally change the optimum parameters for measuring the trend. There is more than one route to your favorite restaurant, but the optimum route will depend on how traffic patterns change over time, on roads that are added or deleted, and on whether traffic lights get installed.

Unsubscribed Premium Strategy: When you view an unsubscribed Premium Strategy, such as this one (click here), you will note the yellow text on the relative strength bar chart that says "Unsubscribed Strategy — 90-day-old trend data." If you are using a Free Strategy, or you have a paid subscription, then this message will not appear and the green bar trends will be current. You may freely evaluate any unsubscribed Premium Strategy and build and evaluate your own Custom Strategies without paying a subscription fee, but the green trend bar chart and the Buy/Sell information will not be current.

• StormGuard ^TM Indicator
The seven StormGuard icons shown in the figure to the right are used on AlphaDroid charts to indicate the general state of the market. When there is a market storm, AlphaDroid's StormGuard algorithm will override the normal selection of the best fund in the Strategy to instead provide a trade signal to move your funds to a safe money market fund or Bear Market Strategy. This is called asset class rotation. The StormGuard Indicator is located on the right side of the title bar of the chart, as shown in the figure to the right. It includes a numerical value which appears to the left of the icon. When the StormGuard Indicator value goes negative the Strategy will produce a Trade Alert (according to the rules below) to indicate that you should move to the safety of a money market fund or Bear Market Strategy. (Note: The default setting is for StormGuard to use the generic symbol $CASH in the Buy/Sell fields to mean "your favorite money market fund." 
 

The character of the StormGuard Indicator is shown in the chart to the right. Its value will typically range from about -4% to +4%. The value is calculated daily using a proprietary algorithm that may be reasonably described as a differential multi-order exponential moving average on a basket of broad market indicators. It is a balanced optimization design that simultaneously reduces whip-saw losses from market dip knee-jerk reactions by not responding too quickly, and minimize the crippling losses of long-duration market storms by not reacting too slowly.   For practical examples of real world events, please check out these two articles:

  •  Japan: A Short Market Slap, or Economic Malaise?  
  •  Should You be Panic Selling on Bad News?  No!   

Please note that the StormGuard Indicator may be slightly different from one Strategy to the next as can be seen when viewing a few of the charts on the Example Strategies page. This is because each Strategy is separately evaluated to determine the amount of storm protection required according the character of its funds. For example, a Strategy composed of broadly diversified funds will do best if it exits to $CASH as soon as broad market averages start a protracted decline, whereas other Strategies composed of sector funds are likely to have one or more funds doing a bit better than the broad market averages as the market decline commences, and thus should be allowed a little more running room before exiting.

StormGuard Versions:  SG-Std,  SG-AQR  and  SG-Armor The standard version of StormGuard is compliant with the trade hold setting for your Strategy, including the Trade Automatic setting which respects extended hold requirements of certain mutual funds. StormGuard - AQR (asymmetric quick response) is designed to "come out of the hole" quicker following a major market selloff, and does not respect the trade hold rules. StormGuard-Armor is designed for maximum safety and the ultimate in using multiple factors to determine when to exit the market. Please read more about these options at the StormGuard-Options page to better understand the features and benefits of each.

The "Faster Response" False Dilemma is a common investment strategy fallacy leading to hair-trigger "ants in the pants" when the markets look scary, with an ever stronger desire for shorter time constants to tighten up the response. The reason this is a false dilemma is because shorter time constants do not lead to better returns. There actually is an optimum time constant, as described earlier, that balances whipsaw losses from when one reacts too quickly, against major decline losses from when one reacts too slowly. Please consider this carefully as you review the meaning of the white chart (that emerges) in the Trade Signal Stationarity topic on the Sector Rotation Theory page. The reason why month-end trading generally performs better is documented in the Hurst Exponent Fingerprint section of the Sector Rotation Theory page. Additionally, if you have not yet done so, please also read this pertinent article: Should You be Panic Selling on Bad News? No! 

•StormGuard Bear Market Symbol & Strategies
The default setting for StormGuard when a bear market rages is to recommend moving to the safety of $CASH (any money market fund). A major market crash can last more than a year, as it did in both the 2001-02 and 2008-09 bear markets. By specifying something other than $CASH for the Bear Market Symbol in StormGuard's advanced options settings, you can do something more proactive than simply hiding until the sun shines on the market again. To access StormGuard's advanced settings, click the Information icon and the button to show the StormGuard options, which include a text box for entering a Bear Market Symbol. The default setting is $CASH, which means that when StormGuard triggers the BUY symbol will be "$CASH" (representing your favorite money market fund). You can change the Bear Market Symbol to anything you like, including (a) mutual funds, ETFs, (b) special extended history ticker symbols, (c) our ready-made Bear Market Strategies ticker symbols, or (d) the special ticker symbols used for the Strategy-of-Strategies. If you use a Strategy-of-Strategies special ticker symbol, such as AD12, then your Strategy will import the equity curve from your own AlphaDroid Strategy #12 as the bear market Strategy to use when StormGuard triggers.

For example, setting the Bear Market Symbol to "UBT-" means that instead of modeling the StormGuard period by simply going to $CASH, it will instead model the period using the extended history version of UBT, a 2x leveraged long-term treasury ETF. Long-term treasury funds are somewhat negatively correlated to the market and will almost always make a worthy improvement to your Strategy's performance, particularly when paired with SG-AQR. Treasury funds vary considerably according to the period of time their holdings are committed. Ten variations are listed in the extended history ticker symbols section, which are provided to aid in Strategy backtesting through artificially extending their history to include one or both market crashes.  Strategies that are tamer do well with TLT-.  Sector rotation Strategies will likely do better with UBT- or TYD-.  More volatile stocks and leveraged ETF Strategies will likely do best with TYD- or TMF-. Numerous ready-made Bear Market Strategies have been designed to span the range of aggressiveness and address the problem of hindsight selection bias that may cause us to favor use of long-term treasuries based on their recent negative correlation to stocks, even though it has not always been the case.

• Score, CAGR, Alpha, Monthly Wins and Beats
Score: The Score value is found in the left portion of each AlphaDroid chart in a statistics section (right). The Score value combines three measures of Strategy performance into a single value that slightly overweights near-term returns and punishes for excess risk in much the same manner as individuals evaluate overall performance.  The formula includes the average return for all years, the average return for the most recent three years, and the Risk Of Loss value for the Strategy. Both long and short term performance are important, and excess risk reduces the Score. The exact equation is shown below. Higher Scores are better.

       Score = (AllYearReturn + 3YearRetrn / 2) / (40% + RiskOfLoss)

Strategies constructed from general purpose diversified funds typically produce Scores in the 45 to 55 range, while a good selection of sector funds will produce a Score in the 80 to 120 range. If a Strategy has only a short history of about 5 years or less, or does not include at least one major market crash during the life of the Strategy, be cautious of reading too much into the comparative value of the Score.

CAGR: CAGR is the compound annual growth rate and is similar, but different from average return. For example, if a fund earned 20%, -5%, and 35% in three consecutive years, it's average return = (10% -5% +35%)/3 = 13.3%. However, the total return of the fund is (1.2 x .95 x 1.35 -1) = 53.9%, but (1.133 x 1.133 x 1.133 -1) = 45.6%. Thus the simply averaging the annual returns does not produce the correct number in a compounding system. The CAGR corrects that problem and is calculated as (Total Return + 1)^(1/yrs)-1. In this case the total return is 53.9% and the period is three years, thus CAGR = (.593 +1)^(.3333)-1 = 15.455%.

 
Alpha: Alpha is a risk-adjusted measure of the so-called active return on an investment and is expressed as an annualized return percentage that reflects the excess performance over the index that would be predicted by Beta alone.  It is commonly used to assess the performance of active fund managers.  It is because AlphaDroid both reduces volatility during market crashes and has higher than usual returns that AlphaDroid has quite excellent alpha measurements.

An asset has a Beta of zero if its returns vary independently of changes in the market's returns. A positive beta means that the asset's returns generally follow the market's returns, in the sense that they both tend to be above their respective averages together, or both tend to be below their respective averages together. Beta is usually measured with respect to an index, such as the S&P 500. A beta of 1.5 would mean that the asset would have correlated price variations 1.5 times greater than the S&P 500.

Alpha and Beta are respectively the vertical intercept and slope of a 'least squares error' line passing through a plotted set of points where the x coordinate is the excess return of the index over an index fund (usually S&P500) and the y coordinate is the excess return of the fund over that index fund. They are calculated as follows:

     Beta = (n*Sxy - Sx*Sy)/(n*Sxx - Sx*Sx)
    
     Alpha = (Sy - Beta*Sx)/n
    
     where S represents the sum of all x, y, x*x, or x*y data over n samples.

Monthly Wins and Beats: Monthly Wins is simply the percentage of months where the Strategy had a positive return. Monthly Beats is the percentage of months that the Strategy beat the reference. By default, the reference is the dividend adjusted S&P500 index, but in the advanced options for a Strategy you can change the reference to be either the AGG aggregate bond fund index or the average of the Strategy's candidate funds.

• 2 Years Rolling Drawdown: Maximum and Average
AlphaDroid uses the 2-year rolling drawdown method as the basis for calculating both the maximum and average drawdown statistics. This industry preferred method additionally enables the plotting of the drawdown over time to help investors identify problems in Strategy construction. In the chart (right), the value plotted at any specific date represents the maximum drawdown measured from a starting point two years earlier through the two year interval. The Maximum and Average values plotted  are enumerated below the chart. The Strategy (yellow) Max Drawdown plotted is 31%, whereas the reference (white - usually the S&P500) Max Drawdown plotted is 57%. The average values plotted for them are 5.4% and 15% respectively. While it is visually obvious that the painful duration of Max Drawdowns for the reference are much longer, this information is explicitly excluded from the Max Drawdown value and explicitly integrated into the Ave Drawdown values.  

• BornOn Date, Last Edit Date, Backtesting and FWPT
In the statistics section on the left side of AlphaDroid charts (right) are two important dates: The Last Edit date indicates when any material part of the Strategy's definition (funds, trade rule, etc.) was last changed, and the BornOn Date indicates the boundary date between pure backtesting (used to tune the algorithms) and the start of FWPT (Forward-Walk Progressive Tuning). Thus AlphaDroid charts inherently show a combination of optimized backtest performance and forward-walk performance through out-of-sample data. During forward-walk, the Strategy is progressively re-tuned at approximately six month intervals (indicated by the yellow dots on the chart's horizontal axis) using only fund data that precedes the re-tuning date. 

The BornOn Date cannot be set for any less than five years from the starting date of the Strategy to help ensure it has suitable data for its initial tuning. Ideally the backtesting period  includes at least one market crash and one bull market period to provide the best opportunity for the algorithm to evaluate the character both market types. In view of these considerations, it is otherwise generally recommended to push the BornOn Date back as far as possible to see if the algorithm can walk forward in time through the out-of-sample data in a reliable manner. Failure to walk forward in time well generally does not mean that the BornOn Date should be moved ahead. What it means is that you have one or more candidate funds that does not transfer leadership to other funds in an orderly manner. Such funds often have a character including strong ramps up followed by sharp drops. The cure is to eliminate such funds because they do not play will with others. It should additionally be understood that if the start date of one or more funds occurs after the BornOn Date the algorithm may perform sub-optimally for a period of time while it adjusts to the new character of the Strategy caused by the newly participating funds. Generally it is advisable to specify a BornOn Date that occurs about a year after the new funds start participating.  Please review other important details about FWPT HERE.