ERQMODELS short for ETF Robo Quant Models
ERQ models define a dynamic allocation of ETFS that are designed to systematically achieve good risk adjusted performance relative to a benchmark like the S&P 500. The ETF SPY is used as a proxy for the S&P 500 index. The models pick which ETFS to buy from a preselected list of ETFs. The preselected list will have a unique name. For example, the sectorx ETFs consist of the following list: SPY XLC XLY XLP XLE XLF XLV XLI XLB XLRE XLK XLU GSG TLT AVUV AVDE AVEM QQQ. As the name suggest sectorx is a list of ETFS that span the 11 sectors of the S&P 500 with the addition of commodities GSG, 20 year treasury TLT, the NASDAQ 100 QQQ, as well as the Avantis group of actively managed funds including AVEM for emerging markets, AVDE for international, AVUV - US SMALL CAP VALUE. The model portfolios uses only end of day price data from a list of selected ETFS. Different models may use different quantitative strategies tc make buy and sell descisions. The models consider the Sharpe,Sortino risk adjusted ratios, the Aroon and RSI indicators and several short and long moving averages to determine when to buy and sell the ETFS. Models will usually hold less than 10 ETFS and may consider dynamic correlation thresholds when choosing ETF combinations. The left-side set of buttons will show the models current performance from corresponding starting dates. The performance chart is displayed on the right-side of the buttons. The daily holdings are shown in a scrollable table below the chart. The table may be scrolled back to the day the model starting trading. The new ETF buys are shown with a green background while ETFs sold have a red background. The percentage cumulative perfomance is shown after the date as Preturn for the model portfolio and SPYreturn for the benchmark return.
Model H1 will allocate up to 6 different sector and style ETFs. The dollar amount of each ETF is determined by dividing the current balance by 6. However, other models may choose to weight the allocations using diffrent strategies. If less than six ETFs are currently in the portfolio, the remaining money is allocated to cash and is shown as a percentage of the current balance. The H1 model assumes no trade cost. H1 also does not consider any tax implications and therefore should best be implement in a tax deferred or tax free account. H1 is not trade frequency constrained so although not common, it may sell an ETF after only 1 day. Other model variations, may impose trade frequency contraints or other techniques to minimze "whipsaw" trading."
Table Heading Legend: Pcum = Portfolio Cumulative Return from initial trading date. Pytd = Portfolio Year to Date Return for the corresponding Year SPYcum = SPY cumulative Return from initial trading date. SPYytd = SPY Year to Date Return for corresponding Year
The Pytod=Pcum and SPYytd=SPYcum for the first year.
Any date or part of a date may be filtered by entering the filter text. It is easy to get the table containing the yearly returns for the portfolio and SPY by entering 31-Dec in the filter prompt box at the top left of the table. The Pytd and the SPYytd for each 31-Dec corresponds to the return for that year.
"Optimal solutions to a mathematical problem are often on the boundary of possible solutions. This causes problems in applying mathematics to real world situation as the mathematical model used to describe the situation is often only an approximation. So the mathematical solutions to the model will often be extreme and since the model is only approximate the solution to the model may be far from optimal." Frank Sortino https://pmpt.me/library/