SUMSRM: A New Statistic for the Structural Break Detection in Time Series
نویسنده
چکیده
Abstract. Structural break is one of the important concerns in non-stationary time series prediction. The cumulative sum of square (CUSUMS) statistic proposed by Brown et al (1975) has been developed as a general method for detecting a structural break. To better understand CUSUMS, this paper analyses the relationship among the bias of the break location estimation, pre-break data size and the decay rate of square residual. Our analysis reveals that small pre-break data size or low decay rate will greatly increase the bias of the break location estimation when there is a change of the mean. Based on the analysis, the paper proposes a new statistic SUMSRM to improve the performance of structural break detection and to reduce the bias of break location estimation. Our empirical evidence confirms that our intended design of the new statistic performs better than the CUSUMS statistic when there is a change of mean in the time series.
منابع مشابه
A New Nonlinear Specification of Structural Breaks for Money Demand in Iran
In a structural time series regression model, binary variables have been used to quantify qualitative or categorical quantitative events such as politic and economic structural breaks, regions, age groups and etc. The use of the binary dummy variables is not reasonable because the effect of an event decreases (increases) gradually over time not at once. The simple and basic idea in this paper i...
متن کاملInvestigation of the Impact of Structural Break on the Relationship between Inflation and Inflation Uncertainty in the Turkish Economy
This article examines the relationship between inflation and inflation uncertainty in the Turkish economy in this period 2004:01-2014:12. This relationship is explored in two ways: a) with the effect of structural breaks; b) without the effect of structural breaks. In fact, with regard to the main structural break have occurred over this period, we examine whether structural break has affected ...
متن کاملHerbal plants zoning using target detection algorithms on time-series of Sentinel-2 multispectral images (Amygdalus Scoparia)
Today, medicinal plants have a special place in the economy and health of a society. Due to the natural growth of many of these products, the necessity of zoning them for optimum and optimal utilization seems necessary. Traditional zoning solutions are not efficient due to their low accuracy and speed, therefore a new approach is needed. Remote sensing data have many applications in various fie...
متن کاملA new adaptive exponential smoothing method for non-stationary time series with level shifts
Simple exponential smoothing (SES) methods are the most commonly used methods in forecasting and time series analysis. However, they are generally insensitive to non-stationary structural events such as level shifts, ramp shifts, and spikes or impulses. Similar to that of outliers in stationary time series, these non-stationary events will lead to increased level of errors in the forecasting pr...
متن کاملOn the Detection of Trends in Time Series of Functional Data
A sequence of functions (curves) collected over time is called a functional time series. Functional time series analysis is one of the popular research areas in which statistics from such data are frequently observed. The main purpose of the functional time series is to predict and describe random mechanisms that resulted in generating the data. To do so, it is needed to decompose functional ti...
متن کامل