Hi, I'm posting my abstrak essay for SEAMS-UGM Conference 2015 quarter. I'm looking to improve it so I'll appreciate any forms of thoughtful feedback. I'm not the best writer and I often miss errors in my own grammar so please point out any awkwardly phrased sentences.
Thanks for taking the time to read all of this.
Government bonds is one of the form of investment that many attractive to investors. Investors cannot know with certainty the level of profits gained because the level of profits gained undergo a change from time to time. An advantage over the level of investment bonds expressed in the percentage called with the yield , to know the value of the yield for the next period, investors need to forecast value of yield. This research using two methods namely by handling the outlier and without handling the outlier data on the results of an parameter estimation. Linear and non-linear parameters derived from estimation modeled first with ARIMA. the best model chosen based on the smallest value of AIC and SBC. Residual obtained from ARIMA models tested heteroscedasticity and the results have heteroscedasticity properties. Time series model used in handling the data without outliers using ARIMA models, ARCH-GARCH and VAR (2), while the handling of outliers on the data time series model used is only a VAR model. The best forecasting model based on criteria out sample is ARIMA model and ARCH-GARCH for handling outliers parameters without parameter done while handling outliers is best VAR models VAR (1). Furthermore, the results of forecasting linear and non-linear parameters obtained are used in constructing the yield curve. Yield curve generated by ARIMA models, ARCH-GARCH and VAR (1) is quite good compared to the VAR model (2).
Thanks for taking the time to read all of this.
Government bonds is one of the form of investment that many attractive to investors. Investors cannot know with certainty the level of profits gained because the level of profits gained undergo a change from time to time. An advantage over the level of investment bonds expressed in the percentage called with the yield , to know the value of the yield for the next period, investors need to forecast value of yield. This research using two methods namely by handling the outlier and without handling the outlier data on the results of an parameter estimation. Linear and non-linear parameters derived from estimation modeled first with ARIMA. the best model chosen based on the smallest value of AIC and SBC. Residual obtained from ARIMA models tested heteroscedasticity and the results have heteroscedasticity properties. Time series model used in handling the data without outliers using ARIMA models, ARCH-GARCH and VAR (2), while the handling of outliers on the data time series model used is only a VAR model. The best forecasting model based on criteria out sample is ARIMA model and ARCH-GARCH for handling outliers parameters without parameter done while handling outliers is best VAR models VAR (1). Furthermore, the results of forecasting linear and non-linear parameters obtained are used in constructing the yield curve. Yield curve generated by ARIMA models, ARCH-GARCH and VAR (1) is quite good compared to the VAR model (2).