Missing information or missing data occurs when no data value is stored for any variable in the dataset. It is usually occurred and can have a substantial effect on the decisions. These days the treatment of the missing data is a real issue. In data quality, a severe threat is missing data which have a direct impact on the certainty of what is being presented to the end user. Imputation of missing values from predictive strategies created the idea that the influences accredited have some association with actual data. If these values did not have an association with actual data, it could produce uncertain results. There are different factual strategies of machine learning techniques like neural systems; genetic programming, and data mining strategies, which are utilized to impute such missing perceptions. In data analytics, predictive methodologies are exceptionally attractive because it can prompt better results. If there are missing data values in datasets certain machine algorithms cannot be applied to predict and forecast data. We can use different time series datasets of different organizations listed in the Pakistan Stock Exchange (PSX). The missing values can be discovered in these selected datasets. We have developed algorithms that can fill the missing dates in the datasets because some of the forecasting algorithms did not initially predict with missing values. The datasets with none missing values can be used for the training of some machine learning algorithms. The results based on different datasets of PSX can be used to forecast the next day value of the stock.