Question
What is the best method and methodology of analysing data to be collected on capital structure determinants?
Answer
Quantitative methods would be best to find the determinants of capital structure in the sense that they look at relationships between variables and allow for theories to be tested. Explanatory/independent variables used to test hypotheses include: institution’s asset structure, age, size, risk effective tax rate, and profitability.
A number of methods have been employed in past studies, some of them are:
1) Panel data methods or procedures are mainly used to investigate the determinants of capital structure. A panel data (or longitudinal data) set consist of a time series for each cross-sectional member in the data set. The key feature of panel data that distinguishes from a pooled cross section is that the same cross -sectional units (individuals, firms, or countries) are followed over a given period of time.
1.1) Fixed Effects Model – takes the presence of unobserved heterogeneity into account and divides the error term into one component that captures the variation between the different institutions analysed and one that captures the remaining observations.
1.2) Random Effects Model – assumes that the data being analysed is drawn from a hierarchy of different populations whose differences relate to that hierarchy
2) Multivariate regression analysis – a technique that estimates a single regression model with more than one outcome variable.
3) Factor analysis (such as maximum likelihood, generalized least squares, unweighted least squares) – this is a method of data reduction. It does this by seeking underlying unobservable variables that are reflected in the observed variables.
Before deciding upon a frequency for data, data availability and size of model should be considered. In the broad scheme of things, if you decide to use higher frequency data (quarterly) which is not available as far back as low frequency data (annual) then you may be omitting important long term trends.
