Interest Rate Forecasting Using Regression Analysis Essay Example
Interest Rate Forecasting using Regression Analysis
Introduction
Forecast of interest rates can be done in many different ways, qualitative (surveys, opinion polls) as well as quantitative (reduced form and structural approaches)*
Example of methods in quantitative approaches - Regression method
- Univariate method (e.g. ARIMA) - Vector autogressive models (VAR) - Single equation approaches - Structural systems of simultaneous equations
This paper will focus on the structural approach relying mainly on the Regression Model technique
Advantages of the structural approach:
Rests on economic theory (unlike reduced form methods such as VAR)
Can trace the effects of changes in macroeconomic variables to interest rates (more likely long rates)
Disadvantages of the structural approach
Data not always readily available at the required frequency
To forecast interest rates using macroeconomic variables imply the use of a structural approach of which 2 processes are involved: 1) model building, and 2) forecasting
Model building: model the relationship between interest rate and relevant macro variables as prescribed by economic theories and quantify (estimate) the relationships using an econometric technique
Forecasting use the estimated model and assumptions on explanatory variables to project future values of the interest rate
Literature Review
A Structural approach to interest rate forecasting
Model building
Economic theories: which economic variables could cause interest rate to deviate from equilibrium
Econometric estimation: at which magnitudes would interest rates be affected by changes in economic variables (quantification of economic relationships)
Economic Theory
What is an interest rate?
Cost of capital (price of borrowing money)
Could be different among various players and uses: Thus different measures of interest rate (short-term, long-term, risk-free etc.)
Example
Introduction
Forecast of interest rates can be done in many different ways, qualitative (surveys, opinion polls) as well as quantitative (reduced form and structural approaches)*
Example of methods in quantitative approaches - Regression method
- Univariate method (e.g. ARIMA) - Vector autogressive models (VAR) - Single equation approaches - Structural systems of simultaneous equations
This paper will focus on the structural approach relying mainly on the Regression Model technique
Advantages of the structural approach:
Rests on economic theory (unlike reduced form methods such as VAR)
Can trace the effects of changes in macroeconomic variables to interest rates (more likely long rates)
Disadvantages of the structural approach
Data not always readily available at the required frequency
To forecast interest rates using macroeconomic variables imply the use of a structural approach of which 2 processes are involved: 1) model building, and 2) forecasting
Model building: model the relationship between interest rate and relevant macro variables as prescribed by economic theories and quantify (estimate) the relationships using an econometric technique
Forecasting use the estimated model and assumptions on explanatory variables to project future values of the interest rate
Literature Review
A Structural approach to interest rate forecasting
Model building
Economic theories: which economic variables could cause interest rate to deviate from equilibrium
Econometric estimation: at which magnitudes would interest rates be affected by changes in economic variables (quantification of economic relationships)
Economic Theory
What is an interest rate?
Cost of capital (price of borrowing money)
Could be different among various players and uses: Thus different measures of interest rate (short-term, long-term, risk-free etc.)
Example