Accelerating linear causal model discovery using hoeffding bounds

Efficiently inducing precise causal models accurately reflecting given data sets is the ultimate goal of causal discovery. The algorithms proposed by Dai et al. has demonstrated the ability of the Minimum Message Length (MML) principle in discovering Linear Causal Models from training data. In order to further explore ways to improve efficiency, this paper incorporates the Hoeffding Bounds into the learning process. At each step of causal discovery, if a small number of data items is enough to distinguish the better model from the rest, the computation cost will be reduced by ignoring the other data items. Experiments with data set from related benchmark models indicate that the new algorithm achieves speedup over previous work in terms of learning efficiency while preserving the discovery accuracy.

Lower bounds for stability margin of n-dimensional discrete systems

This paper derives lower bounds for the stability margin of n-dimensional discrete systems in the Roesser’s state space setting. The lower bounds for stability margin are derived based on the MacLaurine series expansion. Numerical examples are given to illustrate the results.

Revisiting the US money demand function: an application of the Lagrange multiplier structural break unit root test and the bounds test for a long-run relationship

In this article, we examine the issue of a levels relationship and stability of the US money demand function over the period 1959:01 to 2004:02. We use the Lagrange multiplier structural break unit root test and the bounds testing approach to a long-run relationship in levels of the variables, namely real money demand, nominal interest rate and real income. We find greater evidence for a long-run relationship in levels and stability of the US money demand function when we use M2 as a proxy for money demand. However, we find little evidence for a long-run relationship between M1 and M2 with their determinants for the recent period, spanning the last decade or so.

Dead man walking : an empirical reassessment of the deterrent effect of capital punishment using the bounds testing approach to cointegration

This paper empirically estimates a murder supply equation for the United States from 1965 to 2001 within a cointegration and error correction framework. Our findings suggest that any support for the deterrence hypothesis is sensitive to the inclusion of variables for the effects of guns and other crimes. In the long run we find that real income and the conditional probability of receiving the death sentence are the main factors explaining variations in the homicide rate. In the short run the aggravated assault rate and robbery rate are the most important determinants of the homicide rate.

Japan's fertility transition : empirical evidence from the bounds testing approach to cointegration

This study investigates the determinants of the fertility rate in Japan over the 1950–2000 period. We use, for the first time in the fertility literature, the bounds testing approach to cointegration. Amongst our key results, we find that, in both the short-run and long-run, the use of contraceptives and abortion have significantly contributed to the fertility decline in Japan. We also find statistically significant evidence that increasing age at marriage in Japan and increasing education level of women have contributed negatively to the fertility transition.

On performance of transform domain adaptive filters with Markov-2 inputs

In this paper, the analysis for the performance of the discrete Fourier transform LMS adaptive filter (DFT-LMS) and the discrete cosine transform LMS adaptive filter (DCT-LMS) for the Markov-2 inputs is presented. To improve the convergence property of the least mean squares (LMS) adaptive filter, the DFT-LMS and DCT-LMS preprocess the inputs with the fixed orthogonal transforms and power normalization. We derive the asymptotic results for the eigenvalues and eigenvalue distributions of the preprocessed input autocorrelation matrices with DFT-LMS and DCT-LMS for Markov-2 inputs. These results explicitly show the superior decorrelation property of DCT-LMS over that of DFT-LMS, and also provide the upper bounds for the eigenvalue spreads of the finite-length DFT-LMS and DCT-LMS adaptive filters. Simulation results are demonstrated to support the analytic results.

The determinants of immigration from Fiji to New Zealand : AN empirical reassessment using the bounds testing approach

This article re-examines Gani's (1998) findings on the determinants of migrant flows from Fiji to New Zealand by employing the bounds testing procedure to cointegration, within an autoregressive distributive lag framework. The main findings are that in the long run all variables are statistically insignificant, although correctly signed with the exception of the unemployment differential. In the short run, in sharp contrast to Gani's (1998) findings, political instability is consistently the most important determinant of migration flows while the standard of living and real wage differentials are statistically insignificant across all specifications.

Modelling immigration flows : an application of the bounds test to Fiji - Australia migration

This paper examines the long-run and short-run determinants of migration from Fiji to Australia between 1972 and 2001 using a human capital framework, which is extended to take account of political instability in Fiji. Our main findings are that in the long run the real wage differential and political instability in Fiji are the main determinants. In the short run, there is some evidence that the wage differential and transport costs are important factors, but this finding is not robust across all specifications. Lagged migration and political instability are the most important determinants in the short run.

The residential demand for electricity in Australia : an application of the bounds testing approach to cointegration

This paper reports estimates of the long- and short-run elasticities of residential demand for electricity in Australia using the bounds testing procedure to cointegration, within an autoregressive distributive lag framework. In the long run, we find that income and own price are the most important determinants of residential electricity demand, while temperature is significant some of the time and gas prices are insignificant. Our estimates of long-run income elasticity and price elasticity of demand are consistent with previous studies, although they are towards the lower end of existing estimates. As expected, the short-run elasticities are much smaller than the long-run elasticities, and the coefficients on the error-correction coefficients are small consistent with the fact that in the short-run energy appliances are fixed.

Componentwise ultimate bounds for positive discrete time-delay systems perturbed by interval disturbances

This paper presents a method to derive componentwise ultimate upper bounds and componentwise ultimate lower bounds for linear positive systems with time-varying delays and bounded disturbances. The disturbance vector is assumed to vary within a known interval whose lower bound may be different from zero. We first derive a sufficient condition for the existence of componentwise ultimate bounds. This condition is given in terms of the spectral radius of the system matrices which is easy to check and allows us to compute directly both the smallest componentwise ultimate upper bound and the largest componentwise ultimate lower bound. Then, by using the comparison method, we extend the obtained result to a class of nonlinear time-delay systems which has linear positive bounds. Two numerical examples are given to illustrate the effectiveness of the obtained results.