Variance Decomposition Approach Research Articles

Sensitivity analysis of heat exchanger design to uncertainties of correlations

Heat exchanger design procedures rely on a series of correlations to estimate heat transfer coefficients, pressure drops, costs, etc. Each correlation is characterized by uncertainties, i.e. it returns an approximate value of the variable that it aims at predicting. This paper studies how sensitive the total cost evaluation is to these uncertainties for shell-and-tube heat exchangers. A variance decomposition approach is used to calculate the total effect of each uncertain variable. Three test cases are studied for which the most influential uncertainties were found to be those on the correlations for evaluating the purchase cost and the shell side heat transfer coefficient. Probability distributions of the total cost are presented and exhibit a large variance of the total cost. The impact of the cost of energy was also investigated, which revealed that uncertainty on energy cost had a smaller total effect than that of other variables.

Understanding the sources of the exchange rate disconnect puzzle: A variance decomposition approach

In this paper, we apply the monetary model to examine the degree to which economic fundamentals can explain large and persistent fluctuations in the spread between the nominal exchange rate and monetary fundamentals i.e. the so-called “exchange rate disconnect puzzle.” We show that deviations from purchasing power parity account for the bulk of the variation in the deviations of the nominal exchange rate from observed monetary fundamentals, whereas the importance of the risk premium becomes more pronounced after the establishment of the European Economic and Monetary Union.

The Dynamic Relationship between Economic Growth, Tourism Activity, and Real Exchange Rate in Indonesia

This study examines the effect of tourism activity on the economic growth in Indonesia for the period of 1984 - 2014. This study applies the vector autoregressive model, along with the Granger causality test and the persistence profile to analyze the dynamic relationship between tourism activity, economic growth, and real exchange rate in Indonesia. To examine the shock response of the variables, this study uses variance decomposition and impulse response function approach. Throughout analysis is performed based on the empirical literature of the tourism-led growth (TLG) hypothesis. The results show that, based on the co-integration test, there is no long-term relationship between tourism activity and economic growth, while the Granger causality test results no evidence of causality between observed variables, except for economic growth and real exchange rate. In addition, by using variance decomposition and impulse response function to analyze the response of each variable. In explaining the shock of tourism activity in Indonesia, economic growth is more important than the real exchange rate. Meanwhile, tourism activity and real exchange rates are equally important in explaining the shock of economic growth. The impulse response function states that the shock of economic growth and real exchange rate has a positive effect on the tourism activity in the short- and long-term. In addition, the shock of tourism activity has a positive effect on economic growth, while the real exchange rate shock has a negative effect on economic growth in Indonesia.

Precipitation intensity–duration–frequency curves for central Belgium with an ensemble of EURO-CORDEX simulations, and associated uncertainties

An ensemble of 88 regional climate model (RCM) simulations at 0.11° and 0.44° spatial resolutions from the EURO-CORDEX project is analyzed for central Belgium to investigate the projected impact of climate change on precipitation intensity–duration–frequency (IDF) relationships and extreme precipitation quantiles typically used in water engineering designs. The rate of uncertainty arising from the choice of RCM, driving GCM, and radiative concentration pathway (RCP4.5 & RCP8.5) is quantified using a variance decomposition technique after reconstruction of missing data in GCM×RCM combinations. A comparative analysis between the historical simulations of the EURO-CORDEX 0.11° and 0.44° RCMs shows higher precipitation intensities by the finer resolution runs, leading to a larger overestimation of the observations-based IDFs by the 0.11° runs. The results reveal that making a temporal stationarity assumption for the climate system may lead to underestimation of precipitation quantiles up to 70% by the end of this century. This projected increase is generally larger for the 0.11° RCMs compared with the 0.44° RCMs. The relative changes in extreme precipitation do depend on return period and duration, indicating an amplification for larger return periods and for smaller durations. The variance decomposition approach generally identifies RCM as the most dominant component of uncertainty in changes of more extreme precipitation (return period of 10years) for both 0.11° and 0.44° resolutions, followed by GCM and RCP scenario. The uncertainties associated with cross-contributions of RCMs, GCMs, and RCPs play a non-negligible role in the associated uncertainties of the changes.

Is There Froth in the Corporate Bond Market?

Applying the variance decomposition approach to corporate credit spreads, I extract time-varying long-run risk premiums on the corporate bond market portfolio that are easy to compare with the predictions of asset pricing models. The expected cash flows from corporate bonds, identified by predicting default and exercise of embedded call options, are slow-moving, while risk premiums are subject to high-frequency fluctuation and countercyclical. The variation in risk premiums on the corporate bond market portfolio is largely consistent with the CAPM benchmark with time-varying risk exposure.

Impact of Financial Development on Trade Balance: An ARDL Cointegration and Causality Approach for Pakistan

This article aims to examine the relationship among financial development, trade balance, exchange rate and inflation by using time series data from 1972 to 2014 for Pakistan. We have tested the unit root properties of variables by using augmented Dickey–Fuller (ADF), Phillips–Perron and breakpoint unit root tests. The autoregressive distributive lag (ARDL) approach is applied to examine the cointegration between variables due to mixed orders of integration between series I(0)/ I(1). The ARDL findings suggested that long-run relationship exists among financial development, trade balance, exchange rate and inflation. Error correction model (ECM) is applied to analyze short-run relationship. The lagged value of the error correction term (ECMt−1) is negative and significant at 1 per cent level of significance. The value of ECMt−1 is −0.91 which states that digression from the short run towards long run is corrected by almost 91 per cent by every year. Financial development, exchange rate and inflation have significant impact on trade balance in the long run. But in the short run, only exchange rate and inflation have statistically significant impact on trade balance. Diagnostic statistics have confirmed the characteristics of model in the short run as well as in the long run. The causal relationship between variables are examined by vector error correction model (VECM) Granger causality and robustness of causal analysis is tested by variance decomposition approach (VDA). The results of VECM have predicted that unidirectional causality from financial development to trade balance exist in the long run. The results of VDA explained that 19 per cent of trade balance is explained by shocks stimulating financial development. Government should enhance financial development by managing lending interest rates to improve trade balance.

What Drives the Cross‐Section of Credit Spreads?: A Variance Decomposition Approach

ABSTRACTI decompose the variation of credit spreads for corporate bonds into changing expected returns and changing expectation of credit losses. Using a log‐linearized pricing identity and a vector autoregression applied to microlevel data from 1973 to 2011, I find that expected returns contribute to the cross‐sectional variance of credit spreads nearly as much as expected credit loss does. However, most of the time‐series variation in credit spreads for the market portfolio corresponds to risk premiums.

What Drives the Cross-Section of Credit Spreads?: A Variance Decomposition Approach

I decompose the cross-sectional variation of the credit spreads for corporate bonds into changing expected returns and changing expectation of credit losses with a model-free method. Using a log-linearized pricing identity and a vector autoregression applied to micro-level data from 1973 to 2011, I find that the expected credit loss component and the excess return component each explains about half of the variance of the credit spreads. Unlike the market-level findings in Gilchrist and Zakrajsek (2012), at the firm level, the expected credit loss is volatile and affects the firms' investment decision more than the expected excess returns.

Signed Spillover Effects Building on Historical Decompositions

The spillover effects of interconnectedness between financial assets is decomposed into both sources of shocks and whether they amplify or dampen volatility conditions in the target market. We use historical decompositions to rearrange information from a VAR which includes sources, direction and signs of effects building on the unsigned forecast error variance decomposition approach of Diebold and Yilmaz (2009). A spillover index based on historical decompositions has simple asymptotic properties, permitting the derivation of analytical standard errors of the index and its components. We apply the methodology to a panel of CDS spreads of sovereigns and financial institutions for the period of 2003-2013 and identify how these entities contribute to global systemic risk.

The role of the firm in worker wage dispersion: an analysis of the Ghanaian manufacturing sector

This paper uses a linked employer-employee dataset from the Ghanaian manufacturing sector to analyze earnings dispersion in Ghana from 1992 to 2003, a period post extensive economic reforms. I find that variance of earnings increased from 1992 to 1998 and decreased thereafter, resembling an inverted u-shaped relationship. I use analysis of variance and variance decomposition approaches to understand the underlying factors that led to such a pattern in earnings inequality. I find that between-firm factors explain this pattern more than within-firm factors. I also find that the mean earnings gap between workers above and below the 90th percentile of income distribution can explain the majority of the initial surge in inequality (61 %) but only explains a very small fraction of the eventual decline (9 %). I run OLS regressions similar to Mincerian equations and decompose the variance components to find that the decline in earnings inequality is consistent with decline in variance of firm-level earnings whereas variance of predicted wage from worker characteristics have increased. I also find suggestive evidence of changing patterns of worker-firm sorting which contributes to the decline in inequality. These patterns however only hold up for private domestic firms and not for foreign-owned firms. JEL Codes: J31, O15

Exchange Rate, Economic Growth and Foreign Direct Investment in Emerging Asian Economies: Fresh Evidence from Long Run Estimation and Variance Decomposition Approach

This paper investigates the potential relationship between exchange rates, economic growth and foreign direct investment (FDI) for emerging Asian economies. We tested a long run relationship through the ARDL bounds testing approach and the robust regression function which incorporated financial development, capital, openness and trade balance as the additional variables of our model. The findings confirm that exchange rate depreciation affects the volume of FDI and promotes growth in the long run. Economic development and inflows are also associated with exchange rates. The causality test also suggested a long run interdependent link between these three macroeconomic factors. The results of the variance decomposition method suggest that the findings are robust and reliable.

The relationship among oil and coal consumption, carbon dioxide emissions, and economic growth in BRICTS countries

This study aims to analyze the relationship between carbon dioxide (CO2) emissions, economic growth, and coal and oil consumption in Brazil, Russia, India, China, Turkey, and South Africa by using the bounds test approach autoregressive distributed lag (ARDL) over the period from 1969 to 2011. According to ARDL analysis results, it is determined short-run and long-run relationships among selected variables. Three long-run estimators: ARDL cointegration, dynamic ordinary least squares, and fully modified ordinary least squares are utilized to test the robustness of the estimation results. The Granger causality and the forecast error variance decomposition approaches indicate the evidence of a causal relation between variables. According to empirical results, there are the evidence of a uni-directional Granger causality from real gross domestic product (GDP) to carbon dioxide (CO2) emissions in analyzed countries, uni-directional causality from coal consumption to carbon dioxide (CO2) emissions, uni-directional causality from oil consumption to carbon dioxide (CO2) emissions in China, India, Turkey, and South Africa and bi-directional causality in Brazil and Russia. Meanwhile, there is bidirectional causality from GDP to coal consumption, from coal consumption to oil consumption for Brazil, Russia, China, Turkey, and South Africa. India's causality results reveal a bidirectional causality from GDP to coal consumption, oil consumption, and carbon dioxide (CO2) emissions, and the results of forecast error variance supported the results of the causality test.

Imprecise probabilistic estimation of design floods with epistemic uncertainties

AbstractAn imprecise probabilistic framework for design flood estimation is proposed on the basis of the Dempster‐Shafer theory to handle different epistemic uncertainties from data, probability distribution functions, and probability distribution parameters. These uncertainties are incorporated in cost‐benefit analysis to generate the lower and upper bounds of the total cost for flood control, thus presenting improved information for decision making on design floods. Within the total cost bounds, a new robustness criterion is proposed to select a design flood that can tolerate higher levels of uncertainty. A variance decomposition approach is used to quantify individual and interactive impacts of the uncertainty sources on total cost. Results from three case studies, with 127, 104, and 54 year flood data sets, respectively, show that the imprecise probabilistic approach effectively combines aleatory and epistemic uncertainties from the various sources and provides upper and lower bounds of the total cost. Between the total cost and the robustness of design floods, a clear trade‐off which is beyond the information that can be provided by the conventional minimum cost criterion is identified. The interactions among data, distributions, and parameters have a much higher contribution than parameters to the estimate of the total cost. It is found that the contributions of the various uncertainty sources and their interactions vary with different flood magnitude, but remain roughly the same with different return periods. This study demonstrates that the proposed methodology can effectively incorporate epistemic uncertainties in cost‐benefit analysis of design floods.

The Impact of Agricultural Exports on Economic Growth in Nigeria

Agriculture is the backbone of Nigeria’s socioeconomic development. This paper investigates the impact of agricultural exports on economic growth in Nigeria using OLS regression, Granger causality, Impulse Response Function and Variance Decomposition approaches. Both the OLS regression and Granger causality results support the hypothesis that agricultural exports- led economic growth in Nigeria. The results, however, show an inverse relationship between the agricultural degree of openness and economic growth in the country. Impulse Response Function results fluctuate and reveal an upward and downward shocks from agricultural export to economic growth in the country. The Variance Decomposition results also show that a shock to agricultural exports can contribute to the fluctuation in the variance of economic growth in the long run. For Nigeria to experience a favourable trade balance in agricultural trade, domestic processing industries should be encouraged while imports of agricultural commodities that the country could process cheaply should be discouraged. Undoubtedly, this measure could drastically reduce the country’s overreliance on food imports and increase the rate of agricultural production for self-sufficiency, exports and its contribution to the economic growth in the country.

Bridging the variance and diversity decomposition approaches to beta diversity via similarity and differentiation measures

Summary There are many concepts and measures of beta diversity and related similarity/differentiation indices. The variance framework (derived from the total variance of a community species abundance matrix) and diversity decomposition (based on partitioning gamma diversity into alpha and beta components) are two major approaches. There have been no bridges/links between the two approaches. Here, we establish a bridge by extending and modifying each approach so that both lead to the same classes of similarity/differentiation measures, which range in the interval [0, 1] and which can be compared across multiple sets of communities. Our extension/modification in each approach is based on the following major differences between the two approaches. (i) In the decomposition approach, a diversity order q that controls sensitivity to species abundances is used, whereas there is no such order involved in the variance approach. (ii) Transformations of raw abundances are typically used in the variance approach, whereas abundances are not transformed in diversity decomposition. (iii) The variance‐based beta for non‐transformed data is implicitly related to (and constrained by) alpha, gamma and total abundance. Namely, the attained maximum value of this beta when communities are completely distinct (no shared species) is not a fixed constant; the maximum varies with alpha, gamma and total abundance. By contrast, the beta component obtained from the multiplicative decomposition is not constrained by alpha, gamma and total abundance. To construct the bridge, we extend the variance of community data to a class of divergence measures (parameterized by an order q) and use normalization to remove these measures' constraints by alpha, gamma and total abundance. The resulting normalized divergence measures are legitimate differentiation measures. In the decomposition approach, we adopt a modified multiplicative decomposition scheme; the resulting beta component can be transformed to quantify compositional similarity/differentiation among communities. Then, the similarity/differentiation measures obtained from the extended variance framework turn out to be identical to those from the modified diversity decomposition, establishing the bridge. Other types of similarity/differentiation measures (e.g. N‐community Bray–Curtis type) and extension to phylogenetic and functional versions are discussed. A real example using corals is given for illustration.

Dynamics of Military Expenditure and Income Inequality in Pakistan

This study investigates the impact of military expenditures on income inequality in Pakistan using data over the period of 1972–2012. For this purpose, we applied the ARDL bounds testing cointegration approach which confirmed the presence of long-run equilibrium relationship between military expenditure and income inequality. Furthermore, empirical analysis indicates that military spending has a positive impact on income inequality. The analysis of Granger causality, Toda–Yamamoto Modified Wald test and variance decomposition approaches confirm the presence of unidirectional causality running from military expenditure to income inequality. The findings of our study suggest that higher military expenditure leads to higher income inequality in Pakistan. Therefore, we advise the policy makers to focus more on the policies which can increase the economic activities in the country and eventually reduce income inequality.

Corporate payout smoothing: A variance decomposition approach

Abstract In this paper, we apply a variance decomposition methodology to quantify the smoothness of corporate payouts. We find that firms use debt and investment to smooth a large fraction of shocks to net income to keep payouts less variable. Specifically, our empirical results show that firms keep the growth of payouts relatively small and stable over time. Furthermore, our findings support theoretical work that demonstrates that the dynamics of investment and debt policy should be jointly modeled with payout policy.

Global sensitivity analysis for large-scale socio-hydrological models using Hadoop

A multi-agent system (MAS) model is coupled with a physically-based groundwater model to understand the declining water table in the heavily irrigated Republican River basin. Each agent in the MAS model is associated with five behavioral parameters, and we estimate their influences on the coupled models using Global Sensitivity Analysis (GSA). This paper utilizes Hadoop-based Cloud Computing techniques and Polynomial Chaos Expansion (PCE) based variance decomposition approach for the improvement of GSA with large-scale socio-hydrological models. With the techniques, running 1000 scenarios of the coupled models can be completed within two hours with Hadoop clusters, a substantial improvement over the 42 days required to run these scenarios sequentially on a desktop machine. Based on the model results, GSA is conducted with the surrogate model derived from using PCE to measure the impacts of the spatio-temporal variations of the behavioral parameters on crop profits and the water table, identifying influential parameters.

What Drives the Housing Markets in China: Rent, Cost of Capital, or Risk Premium of Owning Versus Renting?

We combine the standard Campbell and Shiller (1988) present-value model with the classical user cost of housing model to decompose the rental yield into three components: expected future rent growth, cost of capital and risk premium of owning versus renting. We then apply a quarterly dataset of first-tier cities (Beijing, Shanghai, Guangzhou and Shenzhen) to explore the question of what forces have driven the movement of China’s housing market. Specifically, using the variance decomposition approach to examine quantitatively how much the variation in rental yield comes from the above-mentioned three components. Our results show that the cost of capital is playing a vital role in all the four major cities, while the future rent growth is not a driving force as significant as the cost of capital in the fluctuation of the rental yield. Moreover, we find a meaningful factor -- the risk of owning versus renting a house -- is accounting for a large part of housing market movement in China. It is also worth noting that the risk of renting versus owning a house seems to be rising rapidly in China over the recent years.

CEO influences on firms' strategic actions: A comparison of CEO-, firm-, and industry-level effects

This study investigates the importance of CEO-level effects for firms' strategic actions, compared with the impact of firm- and industry-level effects. Strategic actions reflect the firm's competitive initiatives and choices regarding financial issues and resource allocation. Drawing on strategic leadership theories, the resource-based view, industrial organization economics, and the reconciling approach of managerial discretion, this study proposes that the CEO's influence varies across different categories of strategic actions. By applying a variance decomposition approach to a 20-year sample of 110 firms in 10 industries, the authors reveal that for competitive initiatives, CEO-level effects are associated with the largest amount of variance, together with firm-level effects. Contrary to expectations, firm-level effects are most relevant for financial choices, followed by CEO effects. Resource allocation depends mainly on firm-level effects, rather than on CEO- or industry-level effects.