Weighting For All Calculations
The Weighting Values (Weight #1 (%)...) is used to assign weights to each input, thus allowing each momentum or volatility or Sharpe ratio or Information Ratio to be assigned different weight values. For example say you would like to make a position score depend 25% on 3 month momentum and 25% on 4 month volatility and 50% on 6 month Sharpe Ratio. You would select the appropriate weight values under each section corresponding to the momentum/volatility/sharpe ratio length to allow for this allocation. When calculating the position score or other metric the weight value then determines the final position score based on 25% momentum, 25% volatility, and 50% Sharpe Ratio as you setup the simulation to run.
The Weighting Values (Weight #1 (%)...) is used to assign weights to each input, thus allowing each momentum or volatility or Sharpe ratio or Information Ratio to be assigned different weight values. For example say you would like to make a position score depend 25% on 3 month momentum and 25% on 4 month volatility and 50% on 6 month Sharpe Ratio. You would select the appropriate weight values under each section corresponding to the momentum/volatility/sharpe ratio length to allow for this allocation. When calculating the position score or other metric the weight value then determines the final position score based on 25% momentum, 25% volatility, and 50% Sharpe Ratio as you setup the simulation to run.
Momentum Calculations
Momentum is defined for our tools as the return % over a specific time frame [Momentum = 100 * (Price Today - Price X Months Ago) / Price Today].
Therefore the length of the momentum is how far we lookback in time (in months, where 1 month = 22 days). So the 3 month momentum of a security in % = the price today minus the price 3 months ago divided by the price today. If 'Actual Months' is selected an actual month (1-1 to 2-1) is used instead of the average 22 days per month.
Example: SPY currently has a price of 200 and had a price 3 months ago of 190, TLT has a price of 130 currently and had a price of 120 3 months ago.
SPY 3 Month Momentum = 100 * (Price Today - Price 3 Months Ago) / Price Today = 100 * (200 - 190) / 200 = 5%
TLT 3 Month Momentum = 100 * (Price Today - Price 3 Months Ago) / Price Today = 100 * (130 - 120) / 130 = 7.6923%
Therefore TLT had the higher 3 month momentum, therefore for a rotation strategy TLT would be chosen over SPY for this period.
Therefore the length of the momentum is how far we lookback in time (in months, where 1 month = 22 days). So the 3 month momentum of a security in % = the price today minus the price 3 months ago divided by the price today. If 'Actual Months' is selected an actual month (1-1 to 2-1) is used instead of the average 22 days per month.
Example: SPY currently has a price of 200 and had a price 3 months ago of 190, TLT has a price of 130 currently and had a price of 120 3 months ago.
SPY 3 Month Momentum = 100 * (Price Today - Price 3 Months Ago) / Price Today = 100 * (200 - 190) / 200 = 5%
TLT 3 Month Momentum = 100 * (Price Today - Price 3 Months Ago) / Price Today = 100 * (130 - 120) / 130 = 7.6923%
Therefore TLT had the higher 3 month momentum, therefore for a rotation strategy TLT would be chosen over SPY for this period.
Volatility Calculations
Volatility is defined for our tools as the standard deviation of a securities returns over a specific time frame [Volatility = Standard Deviation of Rate of change of the securities returns].
Therefore the length of the volatility is how far we lookback in time (in months, where 1 month = 22 days). So the 3 month volatility of a security in % = the standard deviation of that securities returns over a 3 month period.
Therefore the length of the volatility is how far we lookback in time (in months, where 1 month = 22 days). So the 3 month volatility of a security in % = the standard deviation of that securities returns over a 3 month period.
Sharpe Ratio Calculations
The Sharpe Ratio is defined for our tools as the average of the daily returns over a specific time frame divided by standard deviation of a securities returns over a specific time frame [Sharpe Ratio = average of Securities X Month Daily Returns / (Standard Deviation of Rate of change of the securities returns over X Months) ^ Volatility Factor].
The Volatility Factor (sometimes called F Factor) is not a part of the original Sharpe Ratio calculation, if this value is set to 1 the calculation will use the default Sharpe Ratio calculation methods. The Volatility Factor can be changed to adjust how important the volatility (standard deviation of the securities returns) is in the calculation of the Sharpe Ratio. If it is set to a low value the volatility value is small therefore the Sharpe Ratio will be closer to a momentum only strategy; if the volatility factor is set high then the strategy takes into account the volatility of the security much more so it is closer to a volatility only strategy.
Therefore the length of the Sharpe Ratio is how far we lookback in time (in months, where 1 month = 22 days). So the 3 month Sharpe Ratio of a security = Average Daily Return over 3 Months / (Volatility over 3 Months ^ F_Factor).
The Volatility Factor (sometimes called F Factor) is not a part of the original Sharpe Ratio calculation, if this value is set to 1 the calculation will use the default Sharpe Ratio calculation methods. The Volatility Factor can be changed to adjust how important the volatility (standard deviation of the securities returns) is in the calculation of the Sharpe Ratio. If it is set to a low value the volatility value is small therefore the Sharpe Ratio will be closer to a momentum only strategy; if the volatility factor is set high then the strategy takes into account the volatility of the security much more so it is closer to a volatility only strategy.
Therefore the length of the Sharpe Ratio is how far we lookback in time (in months, where 1 month = 22 days). So the 3 month Sharpe Ratio of a security = Average Daily Return over 3 Months / (Volatility over 3 Months ^ F_Factor).
Information Ratio Calculations
The Information Ratio is defined for our tools as the returns over a specific time frame divided by standard deviation of a securities returns over a specific time frame [Information Ratio = Securities X Month Return / (Standard Deviation of Rate of change of the securities returns over X Months) ^ Volatility Factor]. So the Information Ratio is essentially the Momentum / Volatility where volatility is raised to a power (volatility factor or f factor).
The Volatility Factor (sometimes called F Factor) is not a part of the original Information Ratio calculation, if this value is set to 1 the calculation will use the default Information Ratio calculation methods. The Volatility Factor can be changed to adjust how important the volatility (standard deviation of the securities returns) is in the calculation of the Information Ratio. If it is set to a low value the volatility value is small therefore the Information Ratio will be closer to a momentum only strategy; if the volatility factor is set high then the strategy takes into account the volatility of the security much more so it is closer to a volatility only strategy.
Therefore the length of the Information Ratio is how far we lookback in time (in months, where 1 month = 22 days). So the 3 month Information Ratio of a security = Momentum over 3 months / Volatility over 3 months ^ F_Factor.
The Volatility Factor (sometimes called F Factor) is not a part of the original Information Ratio calculation, if this value is set to 1 the calculation will use the default Information Ratio calculation methods. The Volatility Factor can be changed to adjust how important the volatility (standard deviation of the securities returns) is in the calculation of the Information Ratio. If it is set to a low value the volatility value is small therefore the Information Ratio will be closer to a momentum only strategy; if the volatility factor is set high then the strategy takes into account the volatility of the security much more so it is closer to a volatility only strategy.
Therefore the length of the Information Ratio is how far we lookback in time (in months, where 1 month = 22 days). So the 3 month Information Ratio of a security = Momentum over 3 months / Volatility over 3 months ^ F_Factor.
Variance Calculations
Variance is defined for our tools as the average of the daily return squared divided by the number of terms in the calculation.
Therefore the length of the variance is how far we lookback in time (in months, where 1 month = 22 days). So the 3 month variance of a security in % = the average of the daily return values ^2 / (3*22)
Therefore the length of the variance is how far we lookback in time (in months, where 1 month = 22 days). So the 3 month variance of a security in % = the average of the daily return values ^2 / (3*22)
Frequency of Updates
The frequency of updates is simply how often the backtester re-checks the conditions to determine if it should rotate out of a position or change the allocation or check if a security is above or below the moving average. So for a monthly update (as shown above), the backtester will only look re-calculate every 1 month at the end of the month and rotate or adjust allocation.
Top N and Top N Keep K - Rotation Tools Only
In the Rotation tools (Simple and Advanced), you may select how many funds the rotation tool invests in (top n), and if the rotation tool should hold onto the funds until it drops a certain distance in the rankings (keep k).
Top N: This is how many funds the rotation tool will select from the list each period and invest in.
Keep K: This is how far a fund has to drop in the position score ranking before it will be sold at which point it will be replaced by the fund with the highest ranking.
Top N: This is how many funds the rotation tool will select from the list each period and invest in.
Keep K: This is how far a fund has to drop in the position score ranking before it will be sold at which point it will be replaced by the fund with the highest ranking.
Cash Filter Calculations
The cash filter may be used on the Rotation and Moving Average Tools as well as the Volatility Target tool. If a fund's price is below the moving average(s) that fund will be replaced with the Cash Filter Fund Ticker entered on this page. The moving average Filter Method may be selected using different types of moving average calculations. The length may be set to values between 1 and 300, and denotes the number of trading days to do a calculation on.
As an example, if the price of the fund SPY is currently 200, and the value of the 100 day moving average is 201 then at the next period update SPY is below the moving average and therefore instead of investing in SPY the backtester will invest in the cash filter fund selected. Alternatively if the fund SPY is currently 200, and the value of the 100 day moving average is 180 then at the next period update SPY is above the moving average and therefore the cash filter will do nothing thus allowing SPY as a valid option to invest in. If any one of the top funds is below the filter it will invest in the cash filter instead of the selected fund.
The Advanced Rotation Tool allows the user to select 2 different moving average lengths, if a fund is below either moving average then the backtester will invest in the Cash Filter Fund Ticker. Keep in mind the condition for how often to check if a fund is below the moving average (and thus switch it out for the cash filter fund) is only evaluated based on the update frequency described above.
As an example, if the price of the fund SPY is currently 200, and the value of the 100 day moving average is 201 then at the next period update SPY is below the moving average and therefore instead of investing in SPY the backtester will invest in the cash filter fund selected. Alternatively if the fund SPY is currently 200, and the value of the 100 day moving average is 180 then at the next period update SPY is above the moving average and therefore the cash filter will do nothing thus allowing SPY as a valid option to invest in. If any one of the top funds is below the filter it will invest in the cash filter instead of the selected fund.
The Advanced Rotation Tool allows the user to select 2 different moving average lengths, if a fund is below either moving average then the backtester will invest in the Cash Filter Fund Ticker. Keep in mind the condition for how often to check if a fund is below the moving average (and thus switch it out for the cash filter fund) is only evaluated based on the update frequency described above.
Adaptive Asset Allocation Weighting Calculations
In tools that offer Adaptive Asset Allocation, the algorithm for finding position sizing is: Current Ticker's Value / Sum of all values
Example of determining weights for a 3 month Sharpe Ratio based Adaptive Asset Allocation:
SPY has a Sharpe ratio of 1.3
MDY has a Sharpe ratio of 1.7
TLT has a Sharpe ratio of 0.3
GLD has a Sharpe ratio of -0.4 [this is rounded to 0 for calculation purposes]
Sum of all Sharpe Values = 1.3 + 1.7 + 0.3 + 0 [rounded to 0 for calculations] = 3.3
Formula: Current Ticker's Value / Sum of all values
SPY Weight = 1.3 / 3.3 = 39.4%
MDY Weight = 1.7 / 3.3 = 51.5%
TLT Weight = 0.3 / 3.3 = 9.1%
GLD Weight = 0 / 3.3 = 0%
Example of determining weights for a 3 month Sharpe Ratio based Adaptive Asset Allocation:
SPY has a Sharpe ratio of 1.3
MDY has a Sharpe ratio of 1.7
TLT has a Sharpe ratio of 0.3
GLD has a Sharpe ratio of -0.4 [this is rounded to 0 for calculation purposes]
Sum of all Sharpe Values = 1.3 + 1.7 + 0.3 + 0 [rounded to 0 for calculations] = 3.3
Formula: Current Ticker's Value / Sum of all values
SPY Weight = 1.3 / 3.3 = 39.4%
MDY Weight = 1.7 / 3.3 = 51.5%
TLT Weight = 0.3 / 3.3 = 9.1%
GLD Weight = 0 / 3.3 = 0%