A new functional relationship model for circular variables

A new functional relationship model for circular variables which is an extended version of a circular-circular regression model proposed by down and mardia (2002) is developed in this paper. both the dependent and independent variables in the model are subjected to errors. the parameters estimation of the model is done by minimizing the negative value of the log-likelihood function. model verification and its behavior are examined using simulation study with random data. we also derived a fisher information matrix of parameters with the variance and covariance in this new model. as an illustration,a wind direction data obtained using two tools,hf radar (x) and anchored buoy (y) is considered. it is shown that the estimator has small bias thus suggesting the superiority of the new proposed model. © 2014 pakistan journal of statistics.