### Table 7. Akaike and Schwarz Criterion tests for multivariate models

"... In PAGE 18: ...05 Estimation by mixed estimation E. Multivariate Vector Autoregression Table7 reports the result s of the AIC and Schwarz criterion tests for the multivariate models. The criteria suggest using a p max of 11 for the un-differenced models and a p max of 8, for the differenced model.... ..."

### Table 4. Akaike and Schwarz Criterion tests for bivariate models

"... In PAGE 16: ... Bivariate Vector Autoregression of CUREXP amp; PCI Further exploration of the income - expenditure interdependencies continues by assessing the outcomes of the bivariate Vector Autoregression model. AIC and Schwarz criterion tests ( Table4 ) suggest that the maximum lags for the un-differenced (log transformed) model extend to quot;t - 15 quot; while the maximum lags for the differenced model... ..."

### Table 2 Six Prior Model Selection Methods

2002

"... In PAGE 4: ... In this case, the dependent variable yi will be binomially distributed with probability g(H92581, H92582, xi) and the number of binomial trials n, so the shape of error function is completely specified by the experimental task. Six representative selection methods currently in use are shown in Table2 . They are the Akaike information criterion (AIC; Akaike, 1973), the Bayesian information criterion (BIC; Schwarz, 1978), the root mean squared deviation (RMSD), the information- theoretic measure of complexity (ICOMP; Bozdogan, 1990), cross-validation (CV; Stone, 1974), and Bayesian model selection (BMS; Kass amp; Raftery, 1995; Myung amp; Pitt, 1997).... In PAGE 4: ...ifferent value of the variance, that is, ei H11011N(0, H9268i2), (i H11005 1,...,N), then the sample size, n, will now be equal to 1 whereas the data size, N, remains unchanged. 4 The RMSD defined in Table2 differs from the RMSD that has often been used in the psychological literature (e.g.... In PAGE 9: ... Despite these similarities, MDL has at least one advantage over BMS: The complexity measure is well understood. As mentioned above, complexity and GOF are not easily disentangled in the integral form of BMS ( Table2 ). In contrast, a clear understanding of the complexity term in MDL is provided by its counterpart in differential geometry, the geometric complexity measure.... ..."

Cited by 8

### Table 4 Posterior model probabilities for models with non-zero posterior support and the six di erent data sets.

"... In PAGE 11: ...ethod adopted by Brooks et al. (1997). 5.3 Posterior Model Probabilities and Model Averaging Table4 provides the posterior model probabilities for di erent models observed during the simulations and the six data sets under consideration. Here we have taken a at prior over all models with less than or equal to four components.... In PAGE 12: ...3-binomial mixture. [ Table4 about here.] Table 5 provides both the corresponding Akaike apos;s information criterion (AIC) and the Bayesian information criterion (BIC).... ..."

### Table 1: Various criteria for the number of components

"... In PAGE 1: ... The choice of the number of components k was made in terms of the increase in the log likelihood, log L, as k was increased sequentially from k = 1. In Table1 , we report for the three real data sets, the value of log L and its modi ed value under the criteria of AIC (Akaike apos;s information criterion), and BIC (Bayesian information criterion). As is well known, regularity conditions do not hold for the usual asymptotic justi cation of these criteria.... In PAGE 2: ...Table1 the assessed P -value lies between 5% and 10%. Performing further bootstrap replications in these cases did not clarify the situation.... ..."

### Table 3. Multiple regression results between recombination rate and sym- metry measure at order 12 (sym), poly(A)/poly(T) [(A)n gt;4 and (T)n gt;4] tract fraction (pApT), CpG fraction (CpG) and GC content fraction (GC)

2005

"... In PAGE 4: ... Symmetry measure always had a negative correlation with GC content and CpG fraction and a positive correlation with poly(A)/poly(T) [(A)n gt;4 and (T)n gt;4] fraction. Multiple regressions were carried out between local recombina- tion rate and symmetry measure, poly(A)/poly(T) [(A)n gt;4 and (T)n gt;4] fraction, CpG fraction and GC content, and the results are summarized in Table3 . In order to capture potential interactions among sequence features, we performed backward stepwise regres- sion with the Akaike information criterion (AIC) for model selec- tion.... In PAGE 4: ... Therefore, the coefficients and P-values were re-calculated. The final models are shown in Table3 . By using these sequence features, we can explain about 20% of the variance of the local recombination rates for mouse, 19% for rat and 49% for human.... ..."

### Table 5: Model speci cation summary for estimation of GAFT duration speci cations: optimized log-likelihood, Hannan-Quinn (HQ), Akaike (AIC), and Schwarz (BIC) model speci cation criteria.

"... In PAGE 21: ... The mean of the distribution is restricted to be 0 and the estimated variance is approximately 5.16 The left panel of Table5 presents optimized log likelihood values for the various speci cations, and information criteria to aid in selection of the number of terms in the discrete mixture. In- terestingly, the model selection criteria for the NPMLE in Table 5 show the standard pattern; the Schwarz (BIC) and Hannan-Quinn (HQ) criteria both select the two-point speci cation, while the Akaike (AIC) opts for a larger three-point mixture model.... In PAGE 21: ...16 The left panel of Table 5 presents optimized log likelihood values for the various speci cations, and information criteria to aid in selection of the number of terms in the discrete mixture. In- terestingly, the model selection criteria for the NPMLE in Table5 show the standard pattern; the Schwarz (BIC) and Hannan-Quinn (HQ) criteria both select the two-point speci cation, while the Akaike (AIC) opts for a larger three-point mixture model. For reasons that will become clear shortly, I will work primarily with the three term NPMLE and SNP speci cations in the remainder of the discussion In contrast to the straightforward interpretation of the probabilities and support in the NPMLE estimates of F , the coe cients of the SNP presented in Table 6 are more di cult to interpret.... In PAGE 22: ... It is also worth noting that the variances of the estimated mixing distribution roughly coincide for the three point of NPMLE and the three term SNP. The casual observation that the M = 3 expansion appears to be the best SNP speci cation is supported by both model selection criteria ( Table5 ) and visual inspection of the empirical Bayes estimates of the mixing distribution.17 As might be expected, the AIC selects a large model containing three terms in the expansion (as does the Hannan-Quinn criterion).... ..."

### Table 3 contains the deviance values,-21nL, and the Akaike Information

### Table 1: Information criterion results

"... In PAGE 5: ...Fonseca 61 As can be seen, in Table1 , the solution consists of two latent classes. It means that the homoge- neity hypothesis was rejected, thus existing two classes perfectly identified, accordingly with the used information criterion.... ..."

### Table 2: Deviance Information Criterion (DIC) Values

1993

"... In PAGE 17: ... Model Diagnostics and Calibration The Deviance Information Criterion (DIC) (Spiegelhalter et al., 2002) shown in Table2 is used to evaluate the different statistical models (where lower DIC values are better). All of the spatial models perform much better than the model with just the Holland mean function.... ..."

Cited by 1