Skip to content

paulajr/CPI_predictive_model

Repository files navigation

CPI_predictive_model

The objective of this project is to develop a model ARIMA (AutoRegressive Integrated Moving Average) with 31 macroeconomic variables that allows estimating how much the consumer price index in Colombia will vary monthly, which will allow generating investment strategies and offers multiple benefits for investors, regulators and companies in the short and medium term for strengthen financial stability and enable better adaptation to economic fluctuation.

image

Most relevant benefits include:

1. Improved Investment Decision-Making:

  • Investors could anticipate inflation trends and adjust their portfolios accordingly.
  • Inflation-sensitive assets, such as Treasury Inflation-Protected Securities (TIPS) or commodities, could be managed more precisely.
  • It would help forecast interest rate directions, impacting fixed income, equities, and derivatives.

2. More Efficient Risk Management

  • Companies and funds could implement more effective hedging strategies against inflationary fluctuations.
  • Inflation-linked derivatives (swaps, options) could be structured to mitigate risks.
  • Banks could optimize credit policies and interest rates based on inflation scenarios.

3. Impact on Monetary Policy and Regulations

  • Central banks and governments could anticipate inflationary effects and adjust monetary policies proactively.
  • Regulators could foresee inflation crises and take corrective measures before they impact markets.

4. Competitive Advantage in the Financial Sector

  • Investment funds and algorithmic traders could use the model for arbitrage or quantitative trading strategies.
  • Financial analysis firms could provide more accurate forecasts and enhance client advisory services.

5. Optimization of Consumption and Market Behavior

  • Consumer goods companies could adjust pricing and production strategies based on inflation expectations.
  • It could predict changes in demand for goods and services, influencing decisions in retail and manufacturing.

In that order of ideas, the reason for creating this trained model from neural networks in Python arose, whose main source of data was taken directly from the Banco de la Republica´s (BanRep) website.

About

No description, website, or topics provided.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors