An EM Algorithm for Estimating the Parameters of the Normal Inverse Gaussian Distribution with Application in Finance
کد مقاله : 1117-FEMATH (R1)
نویسندگان:
1Hanieh Panahi *، 2fatemeh jafari
1عضو هیئت علمی
2دانشجو/دانشگاه آزاد اسلامی لاهیجان
چکیده مقاله:
The most important step in financial risk is to find a good model for the analyses and estimate the risk measures. The Normal Inverse Gaussian (NIG) distribution is the most used tool for the modeling of financial data. The NIG distribution is able to model symmetric and asymmetric distributions with possibly long tails in both directions. Moreover, the NIG distribution possesses a number of attractive theoretical properties, among others its analytical tractability. Moreover choosing the best method for estimating the parameters of distribution is one of the important aspects in statistical viewpoint. One strength of our approach is that we introduce an Expectation–Maximization (EM) algorithm to compute the maximum likelihood estimates of parameters which involves two steps. A data set of the Tehran Stock Exchange index is used to illustrate the proposed results. We also apply the NIG distribution to evaluation of the Tehran Stock Exchange data in Value-at-Risk framework.
کلیدواژه ها:
EM algorithm; Normal Inverse Gaussian; Tehran Stock Exchange; Value-at-Risk.
وضعیت : مقاله برای ارائه به صورت پوستر پذیرفته شده است