Student Name

Yawar KhalidFollow


Master of Science in Computer Science


Department of Computer Science


School of Mathematics and Computer Science (SMCS)

Date of Submission

Spring 2022


Dr. Tariq Mahmood, Professor and Program Coordinator MS (CS) and MS (DS) Programs, Department of Computer Science


TimeLit is a one-stop time-series forecasting solution to facilitate organizations, data analysts, business analysts, data scientists, investment analysts and financial organizations to determine the best performing and most accurate model for forecasting their future time-series data using their historical data.

The web app works with all univariate time-series datasets may it be sales data, customer engagement, market data or stock price data, our web app runs multiple forecasting models and provides accuracy for all models to determine the best performing one for our particular dataset.

Our app has provisions for resampling, smoothening, differencing, and removing seasonality from the time series to aid in better performance and accurate predictions.

The following 8 models have been incorporated into our web app and our library:

  • Holt-Winters Method (Triple Exponential Smoothening)
  • Seasonal Autoregressive Integrated Moving Average (SARIMA)
  • Linear Regression (with lagging)
  • Linear Regression (with lagging & Feature Engineering)
  • Ridge Regression
  • Lasso Regression
  • Xtreme Gradient Boosting (XG Boost)
  • Long Short Term Model (LSTM)

Timelit also visualizes future data for several timestamps for all models so the user can use that data in their decision-making. For versatility, the user also has the option to run every model stated above individually or run them all at once to compare.

We have also developed a python package with the same functionality for more experienced coders to use and fine-tune as per their requirements.

Document Type

Restricted Access

Submission Type

Research Project

Available for download on Monday, June 15, 2026

The full text of this document is only accessible to authorized users.