Approaches to Economic Forecasting

Econometric Models

The application of statistical methods to model relationships among economic variables.

Structural models specify functional relationships among variables based on economic theory. The functional form and parameters of these models are derived from the underlying theory.

Reduced form models can be models that are simply more-compact representations of underlying structural models or can be models that are essentially data driven, with only a heuristic rationale for selection of variables and/or functional forms.

Economic Indicators

Economic indicators are economic statistics published by official agencies and/or private organizations. These indicators contain information on an economy’s recent past activity or its current or future position in the business cycle.

In particular, a leading indicator moves ahead of the business cycle by a fairly consistent time interval. Most analysts focus primarily on leading indicators because they purport to provide information about upcoming changes in economic activity, inflation, interest rates, and security prices. There are also coincident and lagging indicators that move with and after changes in the business cycle. 

The leading indicators can be used individually or as a composite.

 A composite can also be interpreted as a diffusion index by observing the number of indicators pointing toward expansion versus contraction in the economy.

Checklist Approach

Formally or informally, many forecasters consider a whole range of economic data to assess the economy’s future position. Checklist assessments are straightforward but time-consuming because they require continually monitoring the widest possible range of data. It is also the most subjective method of forecasting

Comparing the Approaches

CFA Book 2 Exhibit 4. Economic Forecasting Approaches: Strengths and Weaknesses

Strengths Weaknesses
Econometric Models Approach
Models can be quite robust, with many factors included to approximate reality.

New data may be collected and consistently used within models to quickly generate output.

Delivers quantitative estimates of impact of changes in exogenous variables.
Imposes discipline/consistency on analysis.
Complex and time-consuming to formulate.

Data inputs not easy to forecast.

Relationships not static.

Model may be mis-specified.

May give false sense of precision.
Rarely forecasts turning points well.
Leading Indicator–Based Approach
Usually intuitive and simple in construction.

Focuses primarily on identifying turning points.

May be available from third parties. Easy to track.
History subject to frequent revision.

Overfitted in-sample.

Likely overstates forecast accuracy.

Can provide false signals.
Checklist Approach
Limited complexity.

Flexible – structural changes easily incorporated.

Items easily added/dropped.

Can draw on any information, from any source, as desired.

Breadth – can include virtually any topics, perspectives, theories, and assumptions.
Subjective.

Time-consuming.

Manual process limits depth of analysis.

No clear mechanism for combining disparate information.

May allow use of biased and/or inconsistent views, theories, assumptions.

Table of Contents

Leave a Comment