• Paul Vettukallel
• Ashwini Madhusudan

Stock price prediction is one of the challenging problems in the field of Machine Learning as stock markets remain extremely fluid due to a varying number of external factors which are fleeting in nature. The performance of the Machine Learning (ML) Models largely depends on the features selected to make predictions. Identification and selection of features and the most appropriate ML Model plays a critical role in prediction accuracy and Model versatility across different stocks and stock markets. This study explores Basic Features as well as Technical Indicators for selection of features for stock prices prediction based on market best practices and seeks to identify the best Train-Test-Scaling combination for modeling the time-series OHLCAV (Open-High-Low-Close-AdjClose-Volume) data that achieves optimum test-metric scores for Linear Regression Models. For this study we used the historical stock prices data from yahoo finance and the prediction process is carried out using Linear Regression, Polynomial Regression and ElasticNetCV Models of ML from the Scikit-Learn library of Python. Our findings reveal a reliable process to identify the best performing Train-Test-Scaler combination for Linear Regression models, for stock price prediction among various other insights.

Machine Learning, Stock Price Prediction, Stock Market Prediction, Linear Regression, Polynomial Regression, Elastic Net, Feature Selection, Standardization, Scaling, Train-Test-Scaler Combination

"Stock Data Feature Selection, Standardisation and Finding the Best Train-Test Combination for Stock Price Prediction using Linear Regression", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.11, Issue 3, page no.d147-d160, March-2024, Available :http://www.jetir.org/papers/JETIR2403318.pdf

"Stock Data Feature Selection, Standardisation and Finding the Best Train-Test Combination for Stock Price Prediction using Linear Regression", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.11, Issue 3, page no. ppd147-d160, March-2024, Available at : http://www.jetir.org/papers/JETIR2403318.pdf