Dear NT team,
Re NT8 strategy analyzer's ability to log X best backtest results (a great thing to have anyway). If a backtest involves many iterations, it is not just the top results which are of interest. Data for worst results can be quite valuable; and, obviously, also for median results.
For example, in a backtest with 500,000 iterations, top 100 results are likely to be curve-fitted and the evolution from worst to best is important.
I appreciate that logging median results (or results by quartile) is difficult, but logging X worst results should not be. Whilst not likely to be a priority for the dev team, it would be great if this is added as an option to NT8 sometime.
Kind regards,
Roman
Re NT8 strategy analyzer's ability to log X best backtest results (a great thing to have anyway). If a backtest involves many iterations, it is not just the top results which are of interest. Data for worst results can be quite valuable; and, obviously, also for median results.
For example, in a backtest with 500,000 iterations, top 100 results are likely to be curve-fitted and the evolution from worst to best is important.
I appreciate that logging median results (or results by quartile) is difficult, but logging X worst results should not be. Whilst not likely to be a priority for the dev team, it would be great if this is added as an option to NT8 sometime.
Kind regards,
Roman
Comment