LSTAR Model Research Articles

Inflation Targeting with an Optimal Nonlinear Monetary Rule—The Case Study of Colombia

This article examines whether Banco de la República (Banrep), Colombia’s central bank, has operated under a dual-regime policy framework—one for recessionary periods and another for periods of economic overheating—since adopting inflation targeting (IT) from Q4 2000 to Q4 2019. We modify the canonical New Keynesian inflation model to accommodate an optimal nonlinear monetary rule aligned with a two-regime policy framework. Using a LSTAR model estimated over the study period, with the output gap lagged by three periods as the transition variable, we identify two distinct monetary regimes. Our findings reveal that the smooth transitions between regimes were driven by shifts in Banrep’s preferences related to its loss function, alongside adjustments in the parameters of the aggregate demand and supply curves within the Colombian economy. Notably, we observe that a modified Taylor principle is not met in either identified monetary regime. This suggests that, in this context, IT has been a successful policy framework even without requiring the policy interest rate to respond aggressively to inflation gaps, as the Taylor principle would otherwise dictate.

Comparing revised latent state-trait models including autoregressive effects.

Understanding the longitudinal dynamics of behavior, their stability and change over time, are of great interest in the social and behavioral sciences. Researchers investigate the degree to which an observed measure reflects stable components of the construct, situational fluctuations, method effects, or just random measurement error. An important question in such models is whether autoregressive effects occur between the residuals, as in the trait-state occasion model (TSO model), or between the state variables, as in the latent state-trait model with autoregression (LST-AR model). In this article, we compare the two approaches by applying revised latent state-trait theory (LST-R theory). Similarly to Eid et al. (2017) regarding the TSO model, we show how to formulate the LST-AR model using definitions from LST-R theory, and we discuss the practical implications. We demonstrate that the two models are equivalent when the trait loadings are allowed to vary over time. This is also true for bivariate model versions. The different but same approaches to modeling latent states and traits with autoregressive effects are illustrated with a longitudinal study of cancer-related fatigue in Hodgkin lymphoma patients. (PsycInfo Database Record (c) 2024 APA, all rights reserved).

PERAMALAN HARGA SAHAM DENGAN METODE LOGISTIC SMOOTH TRANSITION AUTOREGRESSIVE (LSTAR) (Studi Kasus pada Harga Saham Mingguan PT. Bank Mandiri Tbk Periode 03 Januari 2011 sampai 24 Desember 2018)

Series such as financial and economic data do not always form a linear model, so a nonlinear model is needed. One of the popular nonlinear models is the Smooth Transition Autoregressive (STAR). STAR has two possible suitable transition function such as logistic and exponential that need to be test to find the appropriate transition function. The purpose of writing this thesis is to determine the LSTAR model, then use the model to predict the stock price of PT Bank Mandiri. This study uses the data of the weekly stock price of PT Bank Mandiri from the period of January 3, 2011 to December 24, 2018 as insample data and the period of January 1, 2019 to December 30, 2019 as outsample data. The research procedure begins with modeling the data with the Autoregressive (AR) process, testing the linearity of the data, modeling with LSTAR, forecasting, and finally evaluating the results of forecasting. Evaluating the results of the forecasting of the weekly share price of PT Bank Mandiri with the STAR model results in the best nonlinear model LSTAR (1,1). This model produces an highly accurate forecasting result with a value of symmetric Mean Square Error (sMAPE) to be 5.12%.Keywords: Nonlinear, Time Series, STAR, LSTAR.

A nonlinear adjustment in real exchange rates under transaction costs hypothesis in developed and emerging countries

This paper tries to empirically examine the exchange rate deviations to its level under the purchasing power parity (PPP) and transaction costs hypotheses using a battery of newly developed nonlinear approaches. To explain the persistent differences in exchange rates,we use the half-life function analysis with quarterly data over the period 1988Q1–2018Q2 for a panel of 23 developed and emerging countries. Our results show that some deviations to PPP relation may be governed by non-linear dynamics. The adjustment toward the fundamental equilibrium seems to depend on the size and the sign of the gap from the PPP equilibrium. The foreign exchange rate adjustment can be modeled by symmetric process with the ESTAR model and asymmetric process with LSTAR model.

Asymmetric sensitivities of house prices to housing fundamentals: evidence from UK regions

Purpose This paper aims to examine asymmetries in the house price cycle and to understand the dynamic of housing prices, incorporating macroeconomic variables at regional and country level, namely, housing affordability, the unemployment rate, mortgage rate and inflation rate. Design/methodology/approach To highlight significant differences in the asymmetric patterns of house prices between regions, the STAR model is adopted. Findings The authors highlight significant differences in the asymmetric patterns of house prices between regions, in which some areas showed asymmetric response over the housing cycle; here the LSTAR model outperforms other models. In contrast, some regions (the South West and the North West) showed symmetric properties in the tails of the cycle; therefore, the ESTAR model was adopted in their case. Practical implications Being limited to a few fundamentals, this study opens an avenue for further research to investigate this dynamic using in addition such demand-supply factors as land supply, construction cost and loans made for housing. These findings can also be used to examine whether other models such as ARIMA, exponential smoothing or artificial neural networks can more accurately forecast housing prices. Originality/value The present paper aims to highlight housing affordability as a cause of asymmetric behaviour in house prices. Put differently, the authors seek to understand the dynamics of housing prices with other fundamentals incorporating macroeconomic variables in regions and country level data as a means of achieving a more concise result.

STAR Modelleri: Kişi Başına GSYİH Büyüme Hızı İçin Bir Uygulama

The linearity assumption has been valid for a long time but most of the economic time series exhibit asymmetric behavior. The number of studies of nonlinear time series studies have increased in application by removing calculation difficulties. Markov regime switching models, TAR and STAR models are examples of nonlinear time series studies. GDP per capita is an important element of economical growth and has been the subject of many studies. This study covering the annual 1961-2017 data examined the nonlinear structure of GDP per capita growth rate for Turkey with STAR models. The serie is stationary according to the unit root test results and the delay length of the model has been found as two. Transition variables has been created by the delays of the dependent variable one by one and the linearity of the model has been tested against its nonlinearity alternative. The first delay has been selected as the transition variable. At the level of significance of 5%, the linearity hypothesis has been rejected. Findings of the study show that LSTAR model type is appropriate for growth. Since the appropriate model is LSTAR, transition variable has been standardized by its standard error. The parameter indicating the smoothness of transition shows that the transition from one regime to another is quick.

Forecasting performance of smooth transition autoregressive (STAR) model on travel and leisure stock index

Travel and leisure recorded a consecutive robust growth and become among the fastest economic sectors in the world. Various forecasting models are proposed by researchers that serve as an early recommendation for investors and policy makers. Numerous studies proposed distinct forecasting models to predict the dynamics of this sector and provide early recommendation for investors and policy makers. In this paper, we compare the performance of smooth transition autoregressive (STAR) and linear autoregressive (AR) models using monthly returns of Turkey and FTSE travel and leisure index from April 1997 to August 2016. MSCI world index used as a proxy of the overall market. The result shows that nonlinear LSTAR model cannot improve the out-of-sample forecast of linear AR model. This finding demonstrates little to be gained from using LSTAR model in the prediction of travel and leisure stock index.

NONLINEAR SMOOTH TRANSITION AUTOREGRESSIVE (STAR)–TYPE MODELLING AND FORECASTING ON MALAYSIA AIRLINES (MAS) STOCK RETURNS

This study aims to apply nonlinear Smooth Transition Autoregressive (STAR)-type model to the Malaysia Airlines (MAS) Stock Returns, which consists of 4450 number of observations. The data taken started from 29th August 1996 until 26th September 2014. Following the STAR strategies by Terasvirta, the diagnostic plots of linear Autoregressive (AR) model revealed that AR (3) model is adequate in modelling the MAS returns series. However, the squared residuals of Autocorrelation Function (ACF) of returns series illustrates a slight presence of correlations in the model, hence the effort to apply nonlinear model was continued. Before proceed to nonlinear STAR modelling, the identification of delay parameter in the second stage of Terasvirta need to be determined. The results of Lagrange Multiplier (LM) tests revealed that delay parameter, d=3 is the best to choose. In addition, the null hypothesis of linearity from LM test is rejected. Furthermore, from the sequence of nested hypothesis of delay parameter, d=3 indicated that LSTAR model is preferred than ESTAR model. Finally, the forecasts and comparison stages was made to compare which models are best performed in forecasting the future series of MAS returns. It proved that LSTAR model performed better in term of forecasting accuracy when compared to ESTAR and AR model.

Specifying Asymmetric STAR models with Linear and Nonlinear GARCH Innovations: Monte Carlo Approach

Economic and finance time series are mostly asymmetric and are expected to be modeled using asymmetrical nonlinear time series models. Smooth Transition Autoregressive (STAR) models: Logistic (LSTAR) and Exponential (ESTAR) are known to be asymmetric and symmetric respectively, and these have gained their popularity in empirical econometric modelling. Estimating only the LSTAR model for economic and finance data may give models that are not well diagnosed. Under non-normal and heteroscedastic innovations, the residuals of these models are estimated using Generalized Autoregressive Conditionally Heteroscedastic (GARCH) models with variants which include linear and nonlinear forms. The small sample properties of these STAR-GARCH variants are yet to be established but these properties are investigated using Monte Carlo (MC) simulation approach. An MC investigation was conducted to investigate the performance of selections of STAR-GARCH models by the classical nonlinear selection approaches. The ARCH(1) and GARCH(1,1) were the linear GARCH while the Logistic Smooth Transition-ARCH (LST-ARCH(1,1)), Logistic Smooth Transition-GARCH (LST-GARCH(1,1)) and Asymmetric Nonlinear Smooth Transition-GARCH (ANST-GARCH(1,1)) were the nonlinear GARCH. The nonlinearity parameter in the variance equations and Autoregressive (AR) parameters were varied along with different sample sizes. With the assumption of normality, the results showed that the selection of LSTAR models were actually affected by the structure of the innovations and this improved as sample size increased. Misspecification tests actually showed that these models cannot be misrepresented in the real sense.

If the United States sneezes, does the world need “pain-killers”?

In this paper we analyze the extent to which the US economy affects international business fluctuations across countries and we ask whether the nonlinear nature of the business cycle affects the degree of co-movement between countries. A multivariate nonlinear LSTAR model is estimated for the GDP cyclical components of China, France, Germany, the UK and the US. This nonlinear framework allows the asymmetries of the business cycle to be captured properly to identify the synchronization behavior across countries. Our results suggest that there is a relevant influence of the US cycle, specifically during recessions.

Forecasting oil prices: Smooth transition and neural network augmented GARCH family models

The study focuses on a new class of nonlinear volatility models based on neural networks and STAR type nonlinearity. Accordingly, LSTAR–LST–GARCH family and LSTAR–LST–GARCH–NN family of models will be evaluated to analyze petrol prices with economic applications. The nonlinear behavior and leptokurtic distribution are discussed in many studies. The study aims proposing augmentation of linear GARCH, fractionally integrated FI–GARCH and Asymmetric Power APGARCH models with LSTAR type nonlinearity modeling. Further, the proposed models will be augmented with neural networks to benefit from well known learning and forecasting capabilities. The multilayer perceptron (MLP) neural network model and LSTAR model have significant similarities in terms of their architecture. The proposed LSTAR–LST–GARCH family and ANN augmented LSTAR–LST–GARCH–MLP models are evaluated for modeling petrol prices. Empirical findings of the study are: (1) Fractionally integrated and asymmetric power improvements among the GARCH family models provide better forecasting capability for petrol prices; better captured long memory and high volatility characteristics of petrol prices. (2). LSTAR–LST–GARCH model family results in even better gains in out-of-sample forecasting. (3) Donaldson and Kamstra (1997) based MLP–GARCH family provided similar results with the LSTAR–LST–GARCH family models. One exception is for MLP–FIGARCH and MLP–FIAPGARCH models; FI and AP augmented models proposed in this study. (4) Volatility clustering, asymmetry and nonlinearity characteristics of petrol prices are best captured with the LSTAR–LST–GARCH–MLP model family. Forecasting capabilities of neural network techniques are promising. Among the evaluated models, the LSTAR–LST–APGARCH–MLP model provided the best performance overall. With a political perspective, in addition to the highly volatile structure, the long memory characteristics of petrol prices requires that the economic policy interventions should be kept at the modest levels to avoid persistent impacts of shocks.

A LSTAR MODEL WITH TWO THRESHOLDS AND ITS APPLICATION TO RMB EXCHANGE RATE FORECAST

A LSTAR MODEL WITH TWO THRESHOLDS AND ITS APPLICATION TO RMB EXCHANGE RATE FORECAST

Is Tunisian Real Effective Exchange Rate Mean Reverting? Evidence from Nonlinear Models

This paper investigates mean reversion in the deviation of Tunisian real effective exchange rate (REER) from its fundamental value. This paper uses the smooth transition auto-regression (STAR) methodology advocated by Granger and Terasvirta (1993) in order to test whether the Tunisian REER is mean reverting over the period 1990:01 to 2010:01. The empirical results show that data support the hypothesis that deviations can be characterized by asymmetric responses towards appreciation and depreciation with the speed of transition between the expansion and contraction regimes being relatively high. The research results have both methodological and practical originality. On the practical side, the main policy implications the paper puts forward are that foreign exchange market participants should adopt the LSTAR model rather than ESTAR model in their attempt to effectively comprehend the behavior of the exchange rates. On the methodological side, this study makes use of robust test for STAR type nonlinearity. This procedure has been deemed immune against outliers and does not need a priori knowledge regarding their presence and/or timing.

Examining nonlinear dynamics of exchange rates and forecasting performance based on the exchange rate parity of four Asian economies

This paper employs smooth transition autoregressive (STAR) models to investigate the nonlinear dynamic behavior of exchange rate deviations based on the exchange rate parity (ERP) theory in four Asian economies: Japan, South Korea, Taiwan and Singapore. In this study, quarterly data from 1978 Q1 through 2007 Q4 are analyzed. The empirical results indicate that the deviations in the exchange rate of all four countries reject the null of linearity. The rate deviations in the Japanese and Korean cases exhibit a dynamic and smoothly symmetric ESTAR type process, while those in the Taiwanese and Singaporean cases match the smoothly asymmetric LSTAR type with respect to depreciating and appreciating regimes. These nonlinear characteristics can be explained by the existence of heterogeneous behavior and asymmetric information among economic agents. Furthermore, the estimation results of a nonlinear least squares (NLS) regression indicate that most of the parameter estimates are significant at the 10 percent level. The forecasted Japanese and Korean rate deviations in the ESTAR model are not superior to those from the AR model, possibly because these two countries experienced a serious fluctuation during the Asian financial crisis that occurred in 1997. However, based on the criterion of the RMSE, the forecasted Taiwanese and Singaporean rate deviations in the LSTAR model outperform those in the AR model.

Balassa-Samuelson Effect in Won/Dollar and Won/Yen Exchange Rates

This paper examines, using various models including a non-linear one, that the Balassa-Samuelson (BS) effect can account for the persistence of deviations from PPP in the long-run movements of won/dollar and won/yen real exchange rates. In test for PPP hypothesis that incorporates the BS effect, using the generalized Johansen’ cointegration method, it is found that a cointegration relationship exists between each of won/dollar and won/yen real exchange rate and the productivity variables of two countries. And in test for PPP hypothesis that incorporates other fundamentals such as cumulative current account balance, foreign exchange reserve, terms of trade as well as productivity differentials, using a behavioral equilibrium exchange rate approach, it is found that a cointegration relationship exists between each of won/dollar and won/yen real exchange rate and all of these fundamentals. However, the plus sign of the estimated coefficient of the productivity differentials variable, which means that domestic productivity improvement produces increase in each of won/dollar and won/yen real exchange rate is not coincident with the result that the BS effect expects theoretically. Finally, in test for PPP hypothesis that incorporates the BS effect, using a non-linear STAR model, it is found that the adjustment process in case of won/dollar real exchange rate from the long-run equilibrium level can be adequately explained by a non-linear LSTAR model. But, the evidence of diagnostic statistics, which shows the existence of autocorrelation of the residuals in most of lags, might suggest the inadequacy of LSTAR model specification.

Linear and nonlinear causality tests in an LSTAR model: wavelet decomposition in a nonlinear environment

In this paper, we use simulated data to investigate the power of different causality tests in a two-dimensional vector autoregressive (VAR) model. The data are presented in a nonlinear environment that is modelled using a logistic smooth transition autoregressive function. We use both linear and nonlinear causality tests to investigate the unidirection causality relationship and compare the power of these tests. The linear test is the commonly used Granger causality F test. The nonlinear test is a non-parametric test based on Baek and Brock [A general test for non-linear Granger causality: Bivariate model. Tech. Rep., Iowa State University and University of Wisconsin, Madison, WI, 1992] and Hiemstra and Jones [Testing for linear and non-linear Granger causality in the stock price–volume relation, J. Finance 49(5) (1994), pp. 1639–1664]. When implementing the nonlinear test, we use separately the original data, the linear VAR filtered residuals, and the wavelet decomposed series based on wavelet multiresolution analysis. The VAR filtered residuals and the wavelet decomposition series are used to extract the nonlinear structure of the original data. The simulation results show that the non-parametric test based on the wavelet decomposition series (which is a model-free approach) has the highest power to explore the causality relationship in nonlinear models.

Testing for efficiency in selected developing foreign exchange markets: An equilibrium-based approach

This paper proposes an alternative way of testing FOREX efficiency for developing countries. The FOREX market will be efficient if fully reflects all available information. If this holds, the actual exchange rate will not deviate significantly from its equilibrium rate. Moreover, the spot rate should deviate from its equilibrium rate by only transitory components (i.e. it should follow a white noise process). This test is applied to three Central and Eastern European Countries — members of the EU. Considering an LSTAR model we find no evidence of nonlinear adjustment in the misalignment series. So, linear unit root tests imply that the Poland/Euro FOREX market is efficient, the Czech/Euro FOREX market is not, while the Slovak/Euro FOREX market is quasi-efficient.

Comparing smooth transition and Markov switching autoregressive models of US unemployment

AbstractLogistic smooth transition and Markov switching autoregressive models of a logistic transform of the monthly US unemployment rate are estimated by Markov chain Monte Carlo methods. The Markov switching model is identified by constraining the first autoregression coefficient to differ across regimes. The transition variable in the LSTAR model is the lagged seasonal difference of the unemployment rate. Out‐of‐sample forecasts are obtained from Bayesian predictive densities. Although both models provide very similar descriptions, Bayes factors and predictive efficiency tests (both Bayesian and classical) favor the smooth transition model. Copyright © 2008 John Wiley & Sons, Ltd.

Forecasting performance of logistic STAR model: An alternative version to the original LSTAR models

This paper proposes an alternative representation of the original version of the Logistic (LSTAR) model. The LSTAR and ESTAR models are frequently used in empirical research. The LSTAR model describes asymmetrical nonlinear adjustment process, while the ESTAR model describes symmetrical nonlinear adjustment process [19] in Liews [10]. To Liews et al., the theoretical assumption that exchange rate adjustment is symmetric makes the LSTAR model inappropriate for modelling exchange rate movements. They claim that this is the reason the LSTAR model has being neglected in the modelling of exchange rate in the past. They proposed the ALSTAR model. This version they noted allows a V-shape symmetric adjustment in exchange rate behaviour. In this paper, we also propose a similar but better alternative to their model. The square logistic smooth transition auto-regressive model (SLSTAR) has a shape exactly similar to the inverted bell-shape of the exponential model which allows a symmetric adjustment in exchange rate behaviour. Simulation studies were carried out to investigate the sample performance of our model. We also demonstrated our model with life data.

Arbitrage Bounds and the Time Series Properties of the Discount on UK Closed-End Mutual Funds

The persistence of discounts on UK (and US) closed-end mutual funds (investment trusts) is well known, both in the academic literature and in the financial sector. In a dataset of 135 UK ITC's, sampled at weekly intervals, it is shown that while prices and net asset values clearly have unit roots, the evidence suggests that discounts are for the most part not stationary, a finding which supports Cheng, Copeland and O'Hanlon (1994) and more recently Gemmill and Thomas (2002). On the other hand, it is shown here that discounts almost certainly do not have a unit root. A possible statistical reconciliation would be a long memory process, and results are presented to show that for most UK investment trusts the discount is well described by a fractionally-integrated process. This type of model has been widely applied to financial markets ever since it was introduced by Granger and Joyeux (1980). In most cases, the ITC discount process appears to have a root of around 0.7, which implies nonstationarity, but reversion to the mean. In other words, ITC discounts have a long run mean level (of around 13%) to which they eventually return, but disturbances are so persistent that reversion may be an extremely slow, protracted process. This conclusion is in some respects unsatisfactory, given that many of the ITC's in the sample are large, frequently traded, and quite liquid stocks. The long run mean discount estimated here is entirely consistent with the Gemmill and Thomas (2002) arbitrage-induced level. But if arbitrage is at all effective, it is difficult to reconcile with such slow adjustment in the aftermath of disturbances. A possible explanation is to be found in the broader arbitrage bounds generated by the threat of open-ending. If the discount process is actually a bounded random walk, then it is shown here that it will mimic a fractionally-integrated process, a conclusion which may well have relevance to a number of other financial markets which have been shown to display apparent long memory behaviour. The implication is that a more appropriate model of the typical ITC discount would take explicit account of the arbitrage forces at work. Specifically, it is argued that the process would be better characterised by a smooth-transition autoregression (STAR), implying that the speed of adjustment is an increasing function of the gap between the current discount and its long run level. Two specific formats are considered here: the exponential STAR (ESTAR) model, which postulates a symmetric adjustment process, and the logistic (LSTAR) model, which allows for asymmetry. The latter has to be a candidate in the present context, because of the possibility that the speed of adjustment to above-average and below-average discounts are not necessarily the same, since the arbitrage involved is different in the two cases and may involve different transaction costs. In the event, tests on 5 ITC's provide clear evidence of nonlinearity in the discount process, but they are inconclusive about whether the ESTAR or LSTAR model better describes the process.