A generalized time series model based on Kumaraswamy distribution to predict double-bounded relative humidity data

In this study, Kumaraswamy seasonal autoregressive moving average (KSARMA) model was developed to predict double-bounded relative humidity time-series data. In the proposed model, we used the conditional maximum-likelihood method to estimate parameters of the model. For the conditional score vector and conditional Fisher information matrix, the closed type expression were derived. This paper conjointly discusses interval estimation, hypothesis testing, model selection and forecasting. We also used a Monte Carlo simulation to evaluate the finite sample performance of conditional likelihood estimators (CMLEs) and white noise test.