Is a simple MA cross-over strategy good enough (statistical analysis)?
I am approaching this question from a statistical perspective. I took data for EURUSD from 2017.04.20 to 2018.09.18, on M15 chart. I use a simple MA(5) crossover with price close, and the market data was detrended before applying the strategy. Buy and Sell signals based on this simple approach led to returns of around 240% and 220%, respectively (yeah, not a typo); p-value for null hypothesis was zero (machine precision level) in each case. Sample distribution turned out to be almost Gaussian (bootstrap method), with a slight tilt to positive mean returns. I assumed 2:1 leverage too, so not risking too much either (1 lot per trade). Entry and exit were decided by MA(5) crossover with price only. No data mining bias yet. I just took MA(5) just as starters. Where to go from here? Maybe for future, or in some currency pairs these days, where trends are hard to develop, this approach might not work. Some adjustment to MA period might help then (and then I will have to consider data mining bias), or some other range-trading strategy, maybe based on an oscillator like RSI or Stochastics (although I don't trust Stochastics anymore). But this straightforward hypothesis test for MA strategy makes forex trading look quite simple, if I daresay it. I was expecting to be terribly disappointed in the efficacy of such a simple strategy. Maybe the price for EURUSD in that time was strongly trending, so MA strategy works so spectacularly. But then what about the effect of detrending and Bootstrap theorem? I create 999 resamples from the one sample, and so the sampling distribution was created over 1000 samples with a sample size of 500+ each. Is that not sufficient?
Is a simple MA cross-over strategy good enough (statistical analysis)?
I am approaching this question from a statistical perspective. I took data for EURUSD from 2017.04.20 to 2018.09.18, on M15 chart. I use a simple MA(5) crossover with price close, and the market data was detrended before applying the strategy. Buy and Sell signals based on this simple approach led to returns of around 240% and 220%, respectively (yeah, not a typo); p-value for null hypothesis was zero (machine precision level) in each case. Sample distribution turned out to be almost Gaussian (bootstrap method), with a slight tilt to positive mean returns. I assumed 2:1 leverage too, so not risking too much either (1 lot per trade). Entry and exit were decided by MA(5) crossover with price only. No data mining bias yet. I just took MA(5) just as starters. Where to go from here? Maybe for future, or in some currency pairs these days, where trends are hard to develop, this approach might not work. Some adjustment to MA period might help then (and then I will have to consider data mining bias), or some other range-trading strategy, maybe based on an oscillator like RSI or Stochastics (although I don't trust Stochastics anymore). But this straightforward hypothesis test for MA strategy makes forex trading look quite simple, if I daresay it. I was expecting to be terribly disappointed in the efficacy of such a simple strategy. Maybe the price for EURUSD in that time was strongly trending, so MA strategy works so spectacularly. But then what about the effect of detrending and Bootstrap theorem? I create 999 resamples from the one sample, and so the sampling distribution was created over 1000 samples with a sample size of 500+ each. Is that not sufficient?
TradingView India. Detrended Price Oscillator (DPO) — Check out the trading ideas, strategies, opinions, analytics at absolutely no cost! — Indicators and Signals A detrended price oscillator is an oscillator that strips out price trends in an effort to estimate the length of price cycles from peak to peak or trough to trough. The indicator may aid in trade ... What is Detrended Price Oscillator, DPO? Detrended Price Oscillator (DPO) is a dawdling, oscillating period tracker. It is developed to detect cyclical changes in fairness by excluding the overriding pattern from the market activity. DPO identifies the cycles in the underlying price action as prices shift beyond and beneath the displaced SMA. Definition. The Detrended Price Oscillator indicator (DPO) is used to remove trend from price. This is done in order to identify and isolate short-term cycles. DPO is not typically aligned with the most current prices. It is offset to the left (the past) which helps to remove current trend. DiNapoli Detrend Oscillator.mq4. Detrended Price Oscillator. Detrended Price Oscillator (DPO) is an indicator for eliminating trends in prices. DPO allows to more easily identify cycles and, based on that, overbought/oversold levels. Detrended Price Oscillator (DPO) indicator is used to isolate short-term cycles, from long-term cycles.
Technical Analysis using the DETREND PRICE OSCILLATOR This video is a brief tutorial about what the DETREND PRICE OSCILLATOR is and how to use it in vertex fx Software. The presentation is given ... The Detrended Price Oscillator indicator (DPO) is a client side VTL Script used to remove trend from price in Vertex Fx Chart. This is done in order to identify and isolate short-term cycles. DPO... Trading forex involves a significant risk of loss. Always do your own due diligence prior to making an investment decision. ... Backtesting The Detrended Price Oscillator - Duration: 9:01. On Top ... Episode 62 DPO: Detrended Price Oscillator Removing Trend from Price to See Observe Stock Cycles - Duration: ... Mastering Trading Forex Reversals Strategy Urban Forex - Duration: 59:22. Stock Trading Tutor (STT) does not accept any liability in respect of any loss or damage arising from or in connection with any use of the information on or accessed through this Channel or Videos ...
