Missing data is a typical problem in science research. For example, in clinical trials, wearable sensors might lose signal due to battery. Errors in measuring instruments often leading to a gap in time series. Naively dropping missing data can remove important information. In this project, we investigate imputation of missing financial times series data in particular stock time series. We analyze a toy problem where we delete a few data points by hand and attempt to impute it through various methods. The goal is to see which methods and what market indicators work best for such a dataset. The completeness of stock data allows us to test how well a model predicts missing data. Analyzing imputation for such time series could therefore yield insight on correlations in international market and the relevant models and market predictors to use for the more practical problem of making forecast in price movements.