Lab · Comparison
A comparison of the four models and the four baselines in a single table, over the same period and with the same measures.
Comparison table by horizon
What it is: For each horizon, the models' hit rate, 95% interval, number of observations and information coefficient; the baselines in grey rows below.
What data feeds it: The evaluation result of the selected run; walk-forward (750 days training, 63 days test, rolling). In each fold the models are re-estimated using only the training data.
What it shows: The performance of the models and the naive rules on the same test days. A hit is defined the same way everywhere: the signal's direction and the sign of the residual return at the end of the horizon are the same. The residual return is the stock's return with the effect of the market (XU100), the sector and USD/TRY removed; it is not the price itself.
What can be concluded from it: If a model's interval overlaps the intervals of the baselines, that model has found nothing. This table is on residual return; for its raw return counterpart see the Performance screen.
Pairwise tests
What it is: A paired comparison of model pairs on the same stock-days.
What data feeds it: The evaluation result of the selected run; McNemar and Diebold–Mariano tests.
What it shows: For each horizon and pair, the verdict: which model is ahead and the p-value, or "cannot be distinguished yet".
What can be concluded from it: A model coming out "ahead" does not show that the difference is economically meaningful; the hit rate difference and the portfolio with costs should be assessed together.
The investment information, comments and recommendations given here are not within the scope of investment advisory services. Investment advisory services are provided under an investment advisory agreement to be signed between a client and brokerage houses, portfolio management companies, or banks that do not accept deposits. The signals here are produced from historical data with statistical models, are shown the same to everyone and are not personalised; they may not suit your financial situation or your risk and return preferences. Therefore, making investment decisions based solely on the information given here may not produce results that meet your expectations.