Modelling of Electrocardiogram Using Autoregressive Moving Average Model and Linear Predictive Coefficient A Comparative Study

This paper investigate ECG prediction based on the Autoregressive model, Autoregressive Moving Average, and Linear Predictive Coefficient model. It is a complicated and tedious process to predict if the data set is stochastic. The autoregressive model is one of the foremost technique for statistical analysis. Being a linear combination, the model is outlined by the coefficients or the weights within the linear arrangement. By using the best fit models, different models can be evaluated against ECG signal to decide which model portrays the details more reliably. Modelling time series is used for forecasting. The most common methods of modelling time series data for the forecasting of stochastic signals are autoregressive process and autoregressive moving average process. Comparative results have been reported by selecting a model in terms of root mean square error according to the best fits findings. This paper concludes that using a time series model has better forecasting results than the predicted values from existing techniques. The best fit using the model was 97.9%.