Thanks. I know I tend to make things more complicated than they have to be. But when I try simple ideas, they don't usually work out well for me. I do forward test by keeping a record of what these strategies predict after their daily runs using the same amount of historical data for each run. Using some real money for trades but not so much that large losses would affect my standard of living makes me take the trades more seriously. And I understand that spending a lot of effort doesn't mean anything useful will come out of it. For example, I spent a lot of time (months) trying various combinations of John Ehlers-style cyclic indicators from his books (mostly porting EasyLanguage code to perl). But these indicators never worked consistently for me.
Well I haven't been carded for a long time. But to satisfy your curiosity, I am 58 years old. And for full disclosure, my shoe size is 11 (U.S.) .
My Edge(s): Patience is a virtue in a BEAR market Fella who wants pretty nurse must be patient Passionate kiss like spider’s web soon leads to undoing of fly https://finance.yahoo.com/news/ther...19-s-epic-stock-market-rebound-163421765.html
For the genetic programming strategy where I only traded SPDR S&P 500 ETF(SPY) longs, the gains from 02/21/2019 through 03/18/2019 were a whopping 0.08 percent. I think I found more money on a street the other day when I went for a walk . This return is a little misleading because it doesn't include a dividend that hasn't been distributed yet. And I made an ordering mistake where I bought ten times more shares than I should have shortly before the market went down and the strategy said not to buy . [Note to self: don't make any financial transaction when you are talking on the phone to your father who is both hard of hearing and has some dementia .] I tried to make up for this mistake by selling some shares early and not buying at other times during this period. According to dividend-adjusted data, the gain for buy at the close of 02/21/2019 and sell at the close of 03/18/2019 would have been significantly better at 1.77 percent. For the k-nearest neighbor strategy, the gain or loss is harder to calculate because it's on multiple ETFs, and the account has other investments mixed with trades from this strategy. I'll have to write something to parse the results of this strategy from my broker's records.
I have bad gas this year so my brother-in-law can't sit next to me trying to 'learn'. Finally some peace.