In this study, the slope parameter of linear structural relationship model is determined by using the proposed robust nonparametric method based on trimmed mean. This method is an upgrade to the nonparametric method that was introduced by Al-Nasser et al. (2005) by employing trimmed mean for all likely paired slopes rather than median slopes. Simulation study and real data were used to compare the proposed method’s performance versus the traditional maximum likelihood method. In the simulation study, based on both methods’ mean square error, it was inferred that the MLE method break down due to the presence of outliers even though its elaborate was not affected when there was no outlier in the data set. Based on the real life examples, it can be concluded that the performance of our proposed method was better in determining the slope parameter and thus provides a good alternative to MLE method when outliers are present.
In this paper, four types of circular probability distribution were used to evaluate which circular probability distribution gives the best fitting for southwesterly Malaysian wind direction data, namely circular uniform distribution, von Mises distribution, wrapped-normal distribution and wrapped-Cauchy distribution. The four locations chosen were Alor Setar, Langkawi, Melaka and Senai. Two performance indicators or goodness of fit tests which are mean circular distance and chord length were used to test which distribution give the best fitting.
Recently, there is strong interest on the subject of outlier problem in circular data. In this paper, we focus on detecting outliers in a circular regression model proposed by Down and Mardia. The basic properties of the model are available including the exact form of covariance matrix of the parameters. Hence, we intend to identify outliers in the model by looking at the effect of the outliers on the covariance matrix. The method resembles closely the COVRATIO statistic for the case of linear regression problem. The corresponding critical values and the performance of the outlier detection procedure are studied via simulations. For illustration, we apply the procedure on the wind data set.
This paper gives a comprehensive discussion on complex regression model by extending the idea of regression model to circular variables. Various aspect have been considered such as the biasness of parameters, error assumptions and model checking. The advantage of this approach is that it allows the use of usual technique available in ordinary linear regression for the regression of circular variables. The quality of the estimates and the feasibility of the approach were illustrated via simulation. The model was then applied to the wave direction data.
In this study, we propose the estimation of the concentration parameter for simultaneous circular functional relationship model. In this case, the variances of the error term are not necessarily equal and the ratio of the concentration parameter λ = is not necessarily 1. The modified Bessel function was expended by using the asymptotic power series and it became a cubic equation of κ. From the cubic equation of κ, the roots were obtained by using the polyroot function in SPlus software. Simulation study was done to study the mean, estimated bias, absolute relative estimated bias, estimated standard error and estimated root mean square error of the estimation of the concentration parameter. From the simulation study, large concentration parameter and sample size show that the estimated concentration parameter has smaller bias. Also, an illustration to a real wind and wave data set is given to show its practical applicability.
Assessment of instructors by students is needed for assessing the teaching quality of a lecturer towards achieving the objectives of a course. This paper aims to examine the techniques used in sentiment analysis for assessing the effectiveness of a lecturer’s or a teacher’s teaching style in the learning process at a university or school. In addition, the effectiveness of sentiment analysis techniques in assisting the teaching evaluation process is also discussed. The challenges for assessing the quality of teaching of National Defence University of Malaysia (UPNM) lecturers are also discussed in this paper. The sentiment analysis technology is capable of analysing views or opinions on a matter, regardless of whether they are positive or negative. Data from the sentiment analysis can be used by specific parties or anyone else to rectify any weakness or to improve any aspect that the user commented on. The purpose of this study is not to find the weakness of the lecturer, but rather the results of this assessment process can be useful to the management for rectifying weaknesses and for improving the teaching process.
Replicated linear functional relationship model is often used to describe
relationships between two circular variables where both variables have error terms and
replicate observations are available. We derive the estimate of the rotation parameter
of the model using the maximum likelihood method. The performance of the proposed
method is studied through simulation, and it is found that the biasness of the estimates
is small, thus implying the suitability of the method. Practical application of the
method is illustrated by using a real data set.
The existence of outliers in any type of data affects the estimation of models’ parameters. To date there are very few literatures on outlier detection tests in circular regression and it motivated us to propose simple techniques to detect any outliers. This paper considered the complex linear regression model to fit circular data. The complex residuals of complex linear regression model were expressed in two different ways in order to detect possible outliers. Numerical example of the wind direction data was used to illustrate the efficiency of proposed procedures. The results were very much in agreement with the results obtained by using the circular residuals of the simple regression model for circular variables.
This paper focuses on detecting outliers in the circular data which follow the wrapped normal distribution. We considered four discordance tests based on M, C, D and A statistics. The cut-off points of the four tests were obtained and the performance of the detection procedures was studied via simulations. In general, we showed that the discordance test based on the A statistic outperforms the other tests in all cases. For illustration, the city of Kuantan wind direction data set was considered.
Forecasting a multiple seasonal data is differ from a usual seasonal data since it contains more than one cycle in a
data. Multiple linear regression (MLR) models have been used widely in load forecasting because of its usefulness in the
forecast a linear relationship with other factors but MLR has a disadvantage of having difficulties in modelling a nonlinear
relationship between the variables and influencing factors. Neural network (NN) model, on the other hand, is a good
model for modelling a nonlinear data. Therefore, in this study, a combination of MLR and NN models has proposed this
combination to overcome the problem. This hybrid model is then compared with MLR and NN models to see the performance
of the hybrid model. RMSE is used as a performance indicator and a proposed graphical error plot is introduce to see the
error graphically. From the result obtained this model gives a better forecast compare to the other two models.