A Trader’s Journey Through Technical Indicators and Statistical Models

Summary

This reflective account traces a trader’s attempts to find reliable market predictions. The author moves from familiar indicators such as MACD, KDJ, and RSI to custom indicator programming, time-series methods including ARIMA and GARCH, neural networks, support vector machines, and wavelets. Across these approaches, the recurring lesson is that sophistication and historical fit did not remove uncertainty or guarantee useful forecasts of future prices.

The account ends with a return to conventional indicators after a conversation with an experienced trader, emphasizing practical judgment in how tools are applied. It is a personal narrative rather than a systematic comparison: it supplies no datasets, measured accuracy, backtests, or details about the final trader’s methods. Its claims about particular models are therefore the author’s impressions, not general evidence that those methods fail or that traditional indicators outperform them. The piece is useful as a caution about overconfidence in increasingly complex models and about confusing in-sample fit with future performance.

Key ideas

• The author reports uncertainty across traditional indicators, statistical models, and machine-learning approaches.
• Strong historical fit from neural networks did not, in the author’s experience, translate into reliable forecasts.
• The account concludes that practical application matters when using familiar indicators.
• The narrative provides personal impressions rather than comparative tests or performance evidence.

Tags

This summary was written by Stratmill's research agent from the original; it is not a copy of the source.