Unobserved Components Model Research Articles

An output gap measure for the euro area: Exploiting country-level and cross-sectional data heterogeneity

This paper proposes a methodology to estimate the euro-area output gap by taking advantage of two types of data heterogeneity. On the one hand, the method uses information on real GDP, inflation, and the unemployment rate for each member state; on the other hand, it jointly considers this information for all the euro-area countries to extract an area-wide output gap measure. The setup is an unobserved components model that theorizes common output trend and cyclical components across euro-area economies in addition to country-specific counterparts. I estimate the model with Bayesian methods using data for the 19 countries of the euro area from 2000:Q1 through 2018:Q3 and perform model comparisons across different specifications of the output trend. The estimation of the model preferred by the data indicates that, because of negative shocks to trend output during the Global Financial Crisis, output was only slightly below potential in that period, but an output gap of about negative 3% emerged during the European debt crisis. At the end of the sample period, output is estimated to be 3.5% above potential output, which is growing at an average annual pace of only 0.5% in recent years.

Growth cycles and business cycles of the Chinese economy through the lens of the unobserved components model

We build upon the most recent development in the literature and estimate a comprehensive set of univariate and bivariate Unobserved Components models to study the varying long-run growth rate and output gap of the Chinese economy for the period of 1952–2017. We find that the long-run growth rate varied substantially, and thus the growth cycles played an important role in the Chinese economy. Specifically, the long run growth rate accelerated since the late 70s, peaked in the early 90s, and since then has continuously declined except for several years right after China's entrance into the WTO. We also find that although the recent 4-trillion-Yuan economic stimulus package helped to sustain the economic growth temporarily it appeared to have accelerated the decline of the long run growth rate. We also find that the potential growth rate is estimated to be slightly above 8% as in 2017, but the output gap has become negative since 2015 and this recession at the business cycle frequency continuously deepened until the end of the sample.

Explaining the standard errors of corruption perception indices

This paper examines the standard errors of two popular indices of corruption perceptions: the Worldwide Governance Indicators’ Control of Corruption (WGI-CC) and Transparency International's Corruption Perception Index (TI-CPI). The standard errors of these indexes stem from the degree of variation across the sources upon which these two aggregate indices are based. In general, standard errors are not associated with country characteristics; this supports the common assumption that differences across surveys are random. There are two exceptions, however. They involve the degree of media freedom in a country and the country's past corruption scores, possibly indicating the use of cognitive heuristics by the assessors who do the ratings. No evidence exists that more diverse countries have greater variation across corruption scores. In comparing the two aggregate measures, we find that the standard errors for TI-CPI are associated with country characteristics in fewer cases than are those for WGI-CC. Finally, our findings raise concerns about the applicability of the WGI-CC's use of the unobserved components model for extracting signals from noise.

Nominal stability over two centuries

I assess the stability of the monetary environment in Switzerland over the past two centuries. In order to control for transitory measurement errors, in particular in nineteenth century data, I use an unobserved-components stochastic-volatility model to extract the permanent trends from several nominal variables. The descriptive analysis of these trends suggests that the current monetary regime, flexible inflation targeting, provided a relatively stable monetary environment. Although the trends are quite stable for the nineteenth century, the estimates are imprecise. We should therefore be cautious when characterizing metal currency regimes as providing a stable monetary environment. A discussion of the results shows that the apparent success of flexible inflation targeting poses new challenges for the implementation of monetary policy because the trend decline in inflation was associated with a trend decline in nominal interest rates.

RETRACTED ARTICLE: A UCM Approach for Forecasting the Seasonal Rainfall Patterns in Coastal Andhra Pradesh, India 1901–2017

The changes in amount and pattern of seasonal rainfall have a significant impact on agriculture and water resources management in Coastal Andhra Pradesh (CAP) of India. This paper presents modeling and forecasting of the seasonal rainfall patterns in CAP using an unobserved components model (UCM) with the hidden components trend, seasonal, cyclical and irregular. The seasonal rainfall data for CAP were provided by the India Meteorological Department and the analysis for the four rainfall seasons, namely, winter, pre-monsoon, monsoon and post-monsoon. The UCM with deterministic level, deterministic trigonometric seasonal, deterministic cycle and stochastic irregular components is selected from the parsimonious models for forecasting the seasonal rainfall patterns in CAP based on the Bayesian information criterion (BIC), significance tests and statistical fit. The model parameters are obtained by using the maximum likelihood method, and the validity of the selected UCM is determined using normal correlation diagnostics and LJung–Box test statistics for residuals. The forecasting of seasonal rainfall patterns for the years 2018–2020 was carried out with the help of the selected UCM. Further, the UCM forecast reveals that winter rainfall will be around 36.4 mm in 2018, 11.3 mm in 2019 and 54.4 mm in 2020; pre-monsoon rainfall will be 91.3 mm in 2018, 135.1 mm in 2019 and 106.4 mm in 2020; monsoon rainfall will be 686.4 mm in 2018, 663.9 mm in 2019 and 652.9 mm in 2020; post-monsoon rainfall will be 286.2 mm in 2018, 267.9 mm in 2019 and 309.1 mm in 2020.

An unobserved component analysis of Malaysia’s house prices

PurposeThis paper aims to investigate Malaysia’s house prices behaviour by decomposing trend, cycle and stochastic component.Design/methodology/approachThe authors perform an unobserved component model of a structural time series and Markov switching model that covers the period 1999Q1 to 2015Q4.FindingsThe results reveal that the variation in house price in Malaysia is best explained by its trend level, with a small role played by the cycle component; this implies the potential for gaining returns on investments in property by investors and households. The results also show that Malaysia’s HPI cycle is between 8 and 17 years which, in relative terms, is twice the length of the growth cycle and the business cycle in the economy. Meanwhile, the overall movement of HPI is forecast to have a marginal price increase up to 2028Q2.Originality/valueAs house prices remained elevated during the year, the house price dynamic is pivotal for understanding the source of changes in house price. With major findings centred on the relationship between house prices and macroeconomic as well as policy variables, little attention has been paid to composing the trend, cycle and seasonal pattern from the house price index, thus understanding the behaviour of house prices’ unobserved components.

Average labour productivity dynamics over the business cycle

An unobserved components model is used to analyse the cyclicality of average labour productivity in Australia since 1980. The model includes a direct estimate of phase shift so that the lead or lag of cycles relative to output can be determined. Average labour productivity is found to be countercyclical with output and employment and lagging the cycle in output by about six quarters. The observations are consistent with a theoretical model in which employment adjustment costs are falling due to lower turnover in employment. A more flexible labour market allows greater variation in employment relative to variation in output which tends to make productivity more countercyclical. The model is also used to determine the relative importance of the intensive and extensive margins of labour supply by examining fluctuations in both employment rates and average hours per worker. The full-time employment rate is significantly procyclical and lagging output as expected. The part-time employment rate is countercyclical, suggesting that there is some substitution of part time for full-time workers during economic downturns. At the intensive margin, average hours are procyclical, but they make a smaller contribution to the cyclicality of productivity than the employment rate.

Exploring Trend Inflation Dynamics in Euro Area Countries

This paper analyzes the inflation processes of twelve Euro Area countries over the period 1984:q1-2017:q4. The stylized features of inflation uncover its changing nature and cross-country heterogeneity, in terms of mean, volatility and persistence. After estimation of a wide array of unobserved components models, we isolate trend inflation rates in a framework that allows for time-varying inflation gap persistence and stochastic volatility in both the trend and transitory components. On average, a sizeable share of overall inflation dynamics is accounted for by movements in the trend. In explaining trend dynamics, we consistently find a significant role for short-term inflation expectations, economic slack, and openness variables. However, the accumulated impacts of these are fairly small, except in certain, sustained episodes. This is of policy relevance since the monetary authority might want to respond to shocks that are prone to affect the inflation trend in order to ensure that long-term inflation expectations remain anchored.

A threshold unobserved components model of housing bubbles: timings and effectiveness of monetary policies

This paper provides fresh insights into the timings and effectiveness of monetary policies in forestalling the recent housing bubble in the USA. It develops a new version of the threshold unobserved components model, which characterizes interrelations between housing price returns and the federal funds rate during 1991–2010. The framework tracks monetary policies timings through threshold regime-switching phenomena. The estimated low-growth regimes are synchronous with the two bubble-like housing busts in the 1990s and 2006–2007, as well as historical downward movements in the interest rate. The threshold indicator, the moving-average housing price return, indicates a counterintuitively positive association between cyclical housing price returns and the interest rate in the housing bust regimes of many metropolitan housing markets. The results suggest that Phoenix, Riverside, San Diego, and Washington are more vulnerable to housing bubbles than other cities due to the weak effectiveness of monetary policies in stabilizing metropolitan housing markets.

Predictability, Real Time Estimation, and the Formulation of Unobserved Components Models

The formulation of unobserved components models raises some relevant interpretative issues, owing to the existence of alternative observationally equivalent specifi cations, differing for the timing of the disturbances and their covariance matrix. We illustrate them with reference to unobserved components models with ARMA(m;m) reduced form, performing the decomposition of the series into an ARMA(m; q) signal, q m, and a noise component. We provide a characterization of the set of covariance structures that are observationally equivalent, when the models are formulated both in the future and the contemporaneous forms. Hence, we show that, while the point predictions and the contemporaneous real time estimates are invariant to the specifi cation of the disturbances covariance matrix, the reliability cannot be identi fied, except for special cases requiring q < m - 1.

Testing for time variation in the natural rate of interest

SummaryThis paper replicates in a wider sense the unobserved components model of Laubach and Williams (Review of Economics and Statistics, 2003, 85, 1063–1070) to estimate the natural rate of interest (NRI) and investigates the role of model uncertainty. A stochastic Bayesian model selection procedure is employed to test the hypothesis of time variation in the NRI against a constant NRI. The model selection confirms time variation in the NRI as a result of changes in potential output growth, but other determinants of the NRI are found constant.

Real and financial cycles: estimates using unobserved component models for the Italian economy

In this paper we examine the empirical features of both the business and the financial cycle in Italy. We employ univariate and multivariate trend-cycle decompositions based on unobserved component models. Univariate estimates highlight different cyclical properties (persistence, duration and amplitude) of real GDP and real credit to the private sector. Multivariate estimates uncover the presence of feedback effects between the real and the financial cycle. In addition, in the most recent period (2015–2016) the multivariate approach highlights a wider output gap than that estimated by the univariate models considered in this paper.

Estimating Trend Inflation Based on Unobserved Components Model: Is It Correlated with the Inflation Gap?

AbstractBuilding on the work of Stock and Watson (2007), this paper empirically shows that a negative correlation between innovations to trend inflation and the inflation gap plays an important role in the dynamics of postwar U.S. inflation. Additional features that we incorporate in our model include regime‐switching inflation gap persistence and association between inflation and inflation uncertainty. The resulting estimate of trend inflation is smooth, and our model provides superior out‐of‐sample forecasts than Stock and Watson's (2007) unobserved components model with stochastic volatility or than Atkeson and Ohanian's (2001) random walk model does.

Spatial-temporal forecasting of tourism demand

This study conducts spatial-temporal forecasting to predict inbound tourism demand in 29 Chinese provincial regions. Eight models are estimated among a-spatial models (autoregressive integrated moving average [ARIMA] model and unobserved component model [UCM]) and spatial-temporal models (dynamic spatial panel models and space-time autoregressive moving average [STARMA] models with different specifications of spatial weighting matrices). An ex-ante forecasting exercise is conducted with these models to compare their one-/two-step-ahead predictions. The results indicate that spatial-temporal forecasting outperforms the a-spatial counterpart in terms of average forecasting error. Auxiliary regression finds the relative error of spatial-temporal forecasting to be lower in regions characterized by a stronger level of local spatial association. Lastly, theoretical and practical implications are provided.This article also launches the Annals of Tourism Research Curated Collection on Tourism Demand Forecasting, a special selection of research in this field.

Trend and Cycle Shocks in Bayesian Unobserved Components Models for UK Productivity

This paper presents a range of unobserved components models to study productivity dynamics in the United Kingdom. We introduce a set of univariate and bivariate models that allow for shocks between the trend and the cycle to be correlated, and use Bayesian sampling techniques to estimate the models. We show that the size of the priors on the trend and cycle shock has an effect on the results, suggesting that a range of priors need to be considered for policy-making purposes. If the prior is set to a smooth trend, then models with little correlation between the trend and cycle shocks are the likeliest to fit the data. On the other hand, if there is a prior belief that the trend shock is allowed to vary relatively freely, the results suggest that there is a negative correlation between trend and cycle shocks to LIK productivity. This is consistent with real-business cycle type narratives, where trend shocks are the main driver of productivity dynamics. Finally, our evidence suggests that the trend productivity growth rate in the UK has been weaker since the financial crisis. There is also a significant positive correlation between shocks to UK trend productivity and those of other advanced economies.

Is there an association between trends in alcohol consumption and cancer mortality? Findings from a multicountry analysis.

The aim of this analysis is to examine long-term trends in alcohol consumption and associations with lagged data on specific types of cancer mortality, and indicate policy implications. Data on per capita annual sales of pure alcohol; mortality for three alcohol-related cancers - larynx, esophageal, and lip, oral cavity, and pharynx; and per capita consumption of tobacco products were extracted at the country level. The Unobservable Components Model was used for this time-series analysis to examine the temporal association between alcohol consumption and cancer mortality, using lagged data, from 17 countries. Statistically significant associations were observed between alcohol sales and cancer mortality, in the majority of countries examined, which remained after controlling for tobacco use (P<0.05). Significant associations were observed in countries with increasing, decreasing, or stable trends in alcohol consumption and corresponding lagged trends in alcohol-related cancer mortality. Curtailing overall consumption has potential benefits in reducing a number of harms from alcohol, including cancer mortality. Future research and surveillance are needed to investigate, monitor, and quantify the impact of alcohol control policies on trends in cancer mortality.

MODELLING COMMODITY FUTURES PRICES: THE CASE OF CRUDE PALM OIL FUTURES

The present paper seeks to fulfil the gap of knowledge in crude palm oil futures (FCPO) by employing unobserved component (UC) methodology. It may perhaps be the first study in Malaysian capital market to examine FCPO prices using UC model. The empirical results indicate that FCPO presents a permanent behaviour. The transitory component exhibits a higher degree of persistence with a periodicity of approximately one year. In addition, the model of the study has proven its ability to capture unobserved characteristics of FCPO’s time series price and has shown a satisfactory performance for the sampled period in terms of fit. This study implicates that understanding the cyclical pattern, duration, and persistence provides important information that would help to create an appropriate strategy to manage risks, smooth and stabilise returns over the FCPO cycle. Such a strategy would ensure the stability and sustainability of the FCPO market.

Expected demand for Electricity Residential sector in Nineveh governorate for the period (2004-2016) using Unobserved Components Model

اىتم ىذا البحث بد ا رسة الطاقة الکیربائیة فی محافظة نینوى بواقع ) 69 ( شی ا رً ولممدة من 4002 ولغایة 4004 ، أظیر التحمیل الاحصائی لمبیانات المدروسة بانیا تحوی عمى التأثی ا رت الموسمیة والاتجاه العام وکذلک العشوائیة، تم استخدام نموذج المکونات غیر العشوائیة ( UCM ) ( Unobserved Components Model ( فی تحمیل السمسمة المدروسة والتنبؤ لمدة مستقبمیة، . إِذ تم التنبؤ لممدة من 4004 ولغایة 4009 وبالاعتماد عمى احصاءات حسن التوفیق لأخطاء التنبؤ، فقد تبین حسن مطابقة النموذج لمبیانات المستخدمة فی البحث.

A class of periodic trend models for seasonal time series

AbstractTrend and seasonality are the most prominent features of economic time series that are observed at the subannual frequency. Modeling these components serves a variety of analytical purposes, including seasonal adjustment and forecasting. In this paper we introduce unobserved components models for which both the trend and seasonal components arise from systematically sampling a multivariate transition equation, according to which each season evolves as a random walk with a drift. By modeling the disturbance covariance matrix we can encompass traditional models for seasonal time series, like the basic structural model, and can formulate more elaborate ones, dealing with season specific features, such as seasonal heterogeneity and correlation, along with the different role of the nonstationary cycles defined at the fundamental and the harmonic frequencies in determining the shape of the seasonal pattern.

What Drives Output Volatility? The Role of Demographics and Government Size Revisited

AbstractThis paper studies the determinants of output volatility in a panel of 22 OECD countries. In contrast to the existing literature, we avoid ad hoc estimates of volatility based on rolling windows, and we account for possible non‐stationarity. Specifically, output volatility is modelled within an unobserved components model where the volatility series is the outcome of both macroeconomic determinants and a latent integrated process. A Bayesian model selection approach tests for the presence of the non‐stationary component. The results point to demographics and government size as important determinants of macroeconomic (in)stability. A larger share of prime‐age workers is associated with lower output volatility, while higher public expenditure increases volatility.