Table 3.
Binomial regression model for spindle likelihood in older subjects
| Estimate (β) | S.E. | Wald χ 2 | df | p | Odds ratio | 95% CI for odds ratio | |
|---|---|---|---|---|---|---|---|
| (Intercept) | −1.35 | 0.097 | 196.398 | 1 | <.001 | 0.26 | 0.21 to 0.31 |
| Lag | −0.36 | 0.15 | 5.98 | 1.00 | .014 | 0.70 | 0.52 to 0.93 |
| Cos | 0.02 | 0.12 | 0.05 | 1.00 | .830 | 1.03 | 0.81 to 1.31 |
| Sin | 0.18 | 0.12 | 2.48 | 1.00 | .115 | 1.20 | 0.98 to 1.50 |
The binomial regression model for spindle likelihood in older subjects shows that the temporal delay between spindles and the auditory click (Lag) is the main factor driving the logistic regression. Lag: sleep spindle (SS) lag; Cos: click phase cosine; Sin: click phase sine. SE: Standard error, df: degrees of freedom for the Wald test statistic to be compared with the chi-square distribution. Values in bold are statistically significance at alpha = 0.05.