This paper gives an explanation for the failure of machine learning models for the prediction of the cases and the other future trends of Covid-19 pandemic. The paper shows that simple Linear Regression models provide high prediction accuracy values reliably but only for a 2-weeks period and that relatively complex machine learning models, which have the potential of learning long-term predictions with low errors, cannot achieve to obtain good predictions with possessing a high generalization ability. It is suggested in the paper that the lack of a sufficient number of samples is the source of the low prediction performance of the forecasting models. To exploit the information, which is of most relevant with the active cases, we perform feature selection over a variety of variables such as the numbers of active cases, deaths, recoveries, and population. Furthermore, we compare Linear Regression, Multi-Layer Perceptron, and Long-Short Term Memory models each of which is used for prediction of active cases together with various feature selection methods. Our results show that the accurate forecasting of the active cases with high generalization ability is possible up to 3 days because of the small sample size of COVID-19 data. We observe that the Linear Regression model has much better prediction performance with high generalization ability as compared to the complex models but, as expected, its performance decays sharply for more than 14-days prediction horizons.