Science and Education Publishing American Journal of Applied Mathematics and Statistics 2333-4576 2 4 2014 06 12 Modeling Volatility under Normal and Student-t Distributional Assumptions (A Case Study of the Kenyan Exchange Rates) 179 184 EN Rotich Titus Kipkoech Department of Mathematics & Computer Science, University of Eldoret, Eldoret, Kenya AJAMS2014241 10.12691/ajams-2-4-1 2014 05 29 2014 06 11 2014 06 12 The predictive performance of two EGARCH{i} models for modeling daily changes in logarithmic exchange rates (log Rt{ii}) are analyzed here. One is based on modeling the data on assumption of normal distribution and the other is based on the standardized student-t distribution. In particular, the (log Rt) of USDKES{iii}, EUROKES{iv} and GBPKES{v} are considered. For each assumption EGARCH is fitted, with varying numbers of parameters, and attempt to replicate the empirical (log Rt) sequence via simulation. Assessing the fit of each model, it is concluded that the families of EGARCH models with t-innovations adequately reflect the empirical nature of the (log Rt) sequence and therefore provides a better prediction model.