Pension Fund Investment Strategy Based On Monte Carlo Simulation Analysis

Unexpected prolongation in life span, changing accounting rules, low interest rates and dwindling stock market returns have exacerbated the pension fund problems.It has become increasingly apparent that many countries are facing a pension crisis of unprecedented magnitude. Therefore, it is vital essential to find a way to selecting the optimal investment portfolio for the pension fund. In this paper, a data mining method would be used to analyze the output of Monte Carlo model and evaluate different portfolios for pension fund management. Firstly, find the optimal set of deterministic variables for approximation the Monte Carlo distribution of projected funding levels from the stochastic economic scenarios generator. Then build the logistic regression model in R program and use decision analysis to evaluate the model. Afterwards, use historical data to develop the time series model by taking seasonal components into account and then estimate some variables in the project year. At last, predict funding ratios of different portfolios and give the right indication in advance. Therefore, analysis could give an early prediction of’success’ as measured by the funding level being over100%at project year.