Analysing time-series of Quandl

Quandl & Rapporter

2013/11/15 06:29:30 PM

Metadata

Analysing S&P 500 Index downloaded from Quandl in 5.24 seconds with the following original description:

GSPC: S&P 500 Index

Variables

This daily dataset contains 16072 rows and 7 columns with the overall number of 112504 records. As the dataset is made of several variables to analyse, please choose and click on one from the below list for futher univariate analysis:

Overview

Until then, let us check a line chart of the dataset:

There can be seen the 6 variables on the vertical axis based on the date shown on the horizontal axis.

References:

Pair-wise cross-correlation

We can compare the above visualized time-series for relationships between those by the cross-correlation function that is basically a simple Pearson correlation estimate when the lag is 0. The negative and positive numbers on the x axis below shows a lag or delay on a daily basis, which means that we are looking for a possible temporal effect between the variables.

References:

Open and High

It seems that the cross-correlation estimate for Open and High is maximum at lag 0 being 1. So it seems that Open and High are correlated.

High and Low

It seems that the cross-correlation estimate for High and Low is maximum at lag 0 being 1. So it seems that High and Low are correlated.

Low and Close

It seems that the cross-correlation estimate for Low and Close is maximum at lag 0 being 1. So it seems that Low and Close are correlated.

Close and Volume

It seems that the cross-correlation estimate for Close and Volume is maximum at lag -23 being 0.75. So it seems that Close and Volume are correlated.

Volume and Adjusted Close

It seems that the cross-correlation estimate for Volume and Adjusted Close is maximum at lag 23 being 0.75. So it seems that Volume and Adjusted Close are correlated.