This project aim to analyze Standard & Poor's 500 historical data in order to:
- Extract meaningful insights from data
- Forecast future S&P 500 Indexes
S&P 500 (Standard and Poor's 500) is a free-float, capitalization-weighted index of the top 500 publicly listed stocks in the US (top 500 by market cap).
Analyzed Dataset is a multi-variate Time Series one that contains records from 1871-01 to 2018-04.
DATASET SIZE: 1,769 record
FEATURES
Date SP500 Dividend Earnings Consumer Price Index Long Interest Rate Real Price Real Dividend Real Earnings PE10 Date Number Number Number Number Number Number Number Number Number DATASET YEAR BUILD: 2018
- SARIMA
- Holt-Winters
- Facebook Prophet
Component |
---|
Python |
Jupyter Notebook |
Pandas |
SciPy |
StatsModels |
Matplotlib |
NumPy |
Kats |
FilterPy |
PyTorch |
- [1] Bisgaard, Kulahci - Time Series Analysis and Forecasting by Example (First Edition)
- [2] Brockwell, Davis - Introduction to Time Series and Forecasting (Third Edition)
- [3] Jason Brownlee - Introduction to Time Series Forecasting With Python
- [4] Taylor et. al. (Facebook Research) - Forecasting at Scale
- [5] Tashman - Out-of sample tests of forecasting accuracy: an analysis and review
- [6] Gooijer et. al. - 25 Years of Time Series Forecasting
- [7] Local Regression
- [8] Cleveland et. al. - STL: A Seasonal-Trend Decomposition Procedure Based on Loess
- [9] Standard and Poor's (S&P) 500 Index Data
- [10] SciPy, python-based ecosystem for mathematics, science, and engineering
- [11] Kats, one stop shop for time series analysis in Python
- @r-scalia Rosario Scalia