• Keine Ergebnisse gefunden

Supplementary Table S6:

N/A
N/A
Protected

Academic year: 2022

Aktie "Supplementary Table S6:"

Copied!
1
0
0

Wird geladen.... (Jetzt Volltext ansehen)

Volltext

(1)

Supplementary Table S6: Bayesian information criterion difference (ΔBIC) for the investigation of potential mediating effects of smoking status on the association between cpRNFLT and the lipid profile.

Sectors Global T TS TI N NS NI

Total cholesterol (mmol/l) -5.6 -4.2 -8.6 -8.0 -2.6 -4.6 -2.6

HDL cholesterol (mmol/l) -6.3 -3.8 -8.9 -8.4 -2.6 -5.4 -2.8

Non-HDL cholesterol (mmol/l) -6.2 -3.9 -8.8 -8.2 -2.9 -5.3 -3.4

LDL cholesterol (mmol/l) -5.6 -4.0 -8.6 -8.1 -2.6 -4.5 -2.7

TG (mmol/l) -5.8 -4.4 -8.7 -8.0 -2.8 -5.2 -3.0

ApoA1 (g/l) -5.8 -3.4 -8.7 -8.2 -2.5 -4.5 -2.0

ApoB (g/l) -6.1 -4.0 -8.8 -8.1 -2.9 -5.4 -3.1

Lp(a) (g/l) -5.2 -4.0 -8.5 -7.9 -2.3 -4.4 -1.8

Supplementary Table S7

Bayesian information criterion difference (ΔBIC) for the investigation of potential mediating effects of smoking status on the association between cpRNFLT and the lipid profile. For each of the six cpRNFL sectors, two different linear regression models were calculated with age, sex, measurement radius, and the respective lipid marker as regressors (model A), as well as an additional model comprising of model A + smoking status (model B). The BIC difference (ΔBIC) was calculated by ΔBIC = BICmodel A - BICmodel B.ΔBIC values are depicted and a ΔBIC > 2 was regarded as statistically relevant. Abbreviations are indicated in Table 1 and 2.

1

Referenzen

ÄHNLICHE DOKUMENTE

In this paper, I consider the HP model from a Bayesian point of view and I show that the HP smoother is the posterior mean of a (conjugate) Bayesian linear regression model that uses

For each of the six cpRNFL sectors, a linear regression model was calculated with age, sex, and measurement radius, high sensitivity C-reactive protein, as well as the

Unstandardized B coefficients and corresponding p values (corrected for multiple testing based on the false discovery rate method) for the respective

In the previous subsection we have seen that the hat matrix for nonparametric kernel regression consists of four building blocks, the type of local estimation (constant or linear),

We show that the asymptotic variance of the resulting nonparametric estimator of the mean function in the main regression model is the same as that when the selection probabilities

нового статуса участник нашей модели руководствуется тем, что при выборе статуса продавца его новая цена должна быть больше средней цены рынка, определенной в результате

In this note, for the case that the disturbances are conditional homoskedastic, we show that a properly re-scaled residual bootstrap procedure is able to consistently estimate

The asymptotic distribution of OLS in stationary stochastic regression models including long memory processes was first examined by Robinson and Hidalgo (1997).. Specifically,