SVM 1-minute OHLCV vs daily acceptable?

AI & Machine learning
1

Course Name: Trading with Machine Learning: Classification and SVM, Section No: 5, Unit No: 3, Unit type: Notebook

Trading Strategy Using Classification SVM requires 1-minute OHLCV of a stock to be effective? Any use cases where this can be effectivily be used with other time frames such as a daily time frame?

2

Hi Luis,



Here are some specific examples of SVM trading strategies that use daily data:



1. Support and resistance breakout strategy: This strategy involves identifying support and resistance levels using SVMs and then entering a trade when the price breaks above or below a key level. To identify support and resistance levels using SVMs, you can train a model on historical price data. The model will learn to identify the areas where the price has previously found support and resistance. Once the model has been trained, you can use it to predict where the price is likely to find support and resistance in the future, and subsequently decide to go long or short.



2. Trend following strategy: This strategy involves identifying trends using SVMs and then entering a trade in the direction of the trend. A trend is a sustained movement in the price of a security. Trends can be either uptrends (when the price is rising) or downtrends (when the price is falling). To identify trends using SVMs, you can train a model on historical price data. The model will learn to identify the direction of the trend. Once the model has been trained, you can use it to predict the direction of the trend in the future.



3. Mean reversion strategy: This strategy involves identifying overbought and oversold conditions using SVMs and then entering a trade in the opposite direction of the trend. To identify overbought and oversold conditions using SVMs, you can train a model on historical price data. The model will learn to identify the areas where the price is likely to become overbought or oversold. Once the model has been trained, you can use it to predict where the price is likely to become overbought or oversold in the future.

When the model predicts that the price is overbought, this is a bearish signal, meaning that the price is likely to fall. When the model predicts that the price is oversold, this is a bullish signal, meaning that the price is likely to rise.



These are just 3 examples of trading strategies using classification SVM that use daily data. There can be several other such strategies.



Hope this helps.



Thanks

3

Sanskriti,



Thank you for your response. This information you provided really helped. I have a better understanding of SVM applicability timeframes.