Skip to content

lhthien09/INDPRO_index

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

12 Commits
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

INDPRO Nowcasting Project

Description:

Nowcasting is a term used to describe short-term forecasting, i.e. one that can relate to the present, near future or recent past. In economics, technology Nowcasting is primarily used to assess the current state of the economy. It is possible by predicting the value of macroeconomic indicators before being shown their official values for the period.

For example, inflation in the United States for November 2022 was 7.1 percent on an annual basis. This value was published by the U.S. Bureau of Labor Statistics 13 December 2022, while independent economists predicted a few weeks earlier a value of 7.3 percent.

For more info: A Machine Learning approach to estimate current GDP growth

Task:

You are a member of PwC Data Analytics, in a team dealing with the subject nowcasting. Next week you have an important meeting with a client representing one of the larger ones Utilities companies in the United States. This client is interested predicting (nowcasting) the industrial production index for the United States (INDPRO). Your goal is to prepare an econometric or machine learning model predicting the INDPRO indicator. To make this task easier for you, your colleague from team, just before his long-awaited winter holiday in a big hurry, prepared a data frame. There are values for the INDPRO indicator from the last 22 years and some other sample economic indices you may find useful INDPRO predictions. Your team manager asked you to summarize your approach in a short, 3-5 slide presentation. He knows there's not much time left until the meeting, but he's suggested that we do in the first place, focus on presenting the model's effectiveness and dependencies between the variables it detected.

alt text

Variables:

  • y: INDPRO
  • x1: MGASUSGULF
  • x2: GS30
  • x3, x4, x5: ???

Results:

The trained models & report can be found at: Presentation

About

INDPRO time series prediction

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published