Private Equity

Only available on StudyMode
  • Download(s) : 78
  • Published : February 11, 2013
Open Document
Text Preview
FIN 527 Alternative Investments|
Private Equity|
Project 1.|
|
Wei Chen, Ye Zhang|
|

|

Part 1. Performance Measurement for Private Equity
a) Summary statistics for venture capital and buyout returns:

Histograms of returns:

The return distribution of venture capital has a kurtosis of 23.25 and a skewness of 3.63, which means it is leptokurtic and skews to the right. It is not close to normal distribution. Mainly due to the high returns in late 1990s during the "Dot-Com" bubble. The return distribution buyouts have a kurtosis of 2.14 and a skewness of -0.48, which close to 3 and 0, respectively. It is close to normal distribution.

b) Regression results:

From the above result, we can see that R2 is 0.1991, which means the movement of market risk premium can only explain 19.91% the movement of venture capital return. And α is 0.0199 and significant at 10% significance level; β is 0.5883 and significant at 1% significance level. Managers of venture capital seem to beat the market.

The R2 is 0.4774. The movement of market risk premium can explain 47.74% the movement of buyout return. In this regression, α is 0.0162 and significant at 1% significance level; β is 0.3811 and significant at 1% significance level. Managers of buyout seem to beat the market. c) Regression results for venture capital and buyouts with 4 lags:

In this model, R2 is 0.3294, which is higher than 0.1991 in (b). This model fits the data better than the model in (b). The alpha is 0.0058 and not statistically significant 10% significance level. The managers do not seem to beat the market. The beta of 1 quarter lag is not significant at 10% significance level; other betas are all statistically significant at 5% significance level. The new market β, which is the sum of the five β, is significantly greater than the market β estimated in (b).

In this model, R2 is 0.6162, which is also higher than the R2 in (b). This model fits the data better....
tracking img