The Product-Limit Survival Fit red triangle menu contains the following options:
Survival Plot
Shows or hides the survival plot, which contains overlaid survival curves for each group.
Failure Plot
Shows or hides the failure plot, which contains overlaid failure curves (proportion failing over time) for each group. A failure plot reverses the vertical axis to show the number of failures rather than the number of survivors. This is useful in reliability analysis.
Plot Options
Contains the following options:
Note: The first seven options (Show Points, Show Kaplan Meier, Show Combined, Show Confid Interval, Show Simultaneous CI, Show Shaded Pointwise CI, and Show Shaded Simultaneous CI) and the last two options (Fitted Survival CI, Fitted Failure CI) pertain to the initial survival plot and failure plot. The other five (Midstep Quantile Points, Connect Quantile Points, Fitted Quantile, Fitted Quantile CI Lines, Fitted Quantile CI Shaded) pertain only to the distributional plots.
Show Points
Shows or hides the points in the Survival Plot and the Failure Plot. Failures appear at the bottom of the steps, and censored observations are indicated by points above the steps.
Show Kaplan Meier
Shows or hides the Kaplan-Meier survival functions for each group in the Survival Plot and the Failure Plot. This option is on by default.
Show Combined
Shows or hides the combined Kaplan-Meier survival functions in the Survival Plot and the Failure Plot.
Show Confid Interval
Shows or hides the 95% pointwise confidence bands for the Kaplan-Meier survival functions in the Survival Plot and the Failure Plot. This option also shows confidence bands for the combined survival functions when the Show Combined option is selected.
Show Simultaneous CI
Shows or hides the 95% simultaneous confidence bands for the Kaplan-Meier survival functions in the Survival Plot and the Failure Plot. This option also shows confidence bands for the combined survival functions when the Show Combined option is selected. Meeker and Escobar (1998, ch. 3) discuss pointwise and simultaneous confidence intervals and the motivation for simultaneous confidence intervals in survival analysis.
Show Shaded Pointwise CI
Shows or hides shaded regions for the 95% pointwise confidence bands for the Kaplan-Meier survival functions in the Survival Plot and the Failure Plot. This option also shows shaded confidence regions for the combined survival functions when the Show Combined option is selected.
Show Shaded Simultaneous CI
Shows or hides shaded regions for the 95% simultaneous confidence bands for the Kaplan-Meier survival functions in the Survival Plot and the Failure Plot. This option also shows confidence bands for the combined survival functions when the Show Combined option is selected. Meeker and Escobar (1998, ch. 3) discuss pointwise and simultaneous confidence intervals and the motivation for simultaneous confidence intervals in survival analysis.
Midstep Quantile Points
Specifies that the modified Kaplan-Meier plotting positions are used in the Exponential Plot, the Weibull Plot, and the LogNormal Plot. These plotting positions are equivalent to taking mid-step positions of the Kaplan-Meier curve, rather than the bottom-of-step positions. This option on by default.
Connect Quantile Points
Shows or hides the lines in the Exponential Plot, the Weibull Plot, and the LogNormal Plot. This option is on by default.
Fitted Quantile
Shows or hides straight-line fits for each group in the Exponential Plot, the Weibull Plot, and the LogNormal Plot. This option is on by default.
Fitted Quantile CI Lines
Shows or hides the 95% confidence bands for each group in the Exponential Plot, the Weibull Plot, and the LogNormal Plot.
Fitted Quantile CI Shaded
Shows or hides shaded regions for the 95% confidence bands for each group in the Exponential Plot, the Weibull Plot, and the LogNormal Plot.
Fitted Survival CI
Shows or hides confidence intervals for each group in the Survival Plot. A set of intervals is plotted for each of the fitted distributions.
Fitted Failure CI
Shows or hides confidence intervals for each group in the Failure Plot. A set of intervals is plotted for each of the fitted distributions.
Exponential Plot
Shows or hides the Exponential Plot, which shows the cumulative exponential failure probability by time for each group. Lines that are approximately linear empirically indicate the appropriateness of using an exponential model for further analysis. For example, in Figure 14.5, the lines for Group 1 and Group 2 in the Exponential Plot are curved rather than straight. This indicates that the exponential distribution is not appropriate for this data. See Exponential, Weibull, and Lognormal Plots and Fits.
Exponential Fit
Shows or hides the Exponential Parameter Estimates table. This option also adds a linear fit to the exponential cumulative distribution function in the Exponential Plot. The θ parameter corresponds to the mean failure time. See Exponential, Weibull, and Lognormal Plots and Fits.
Weibull Plot
Shows or hides the Weibull Plot, which shows the cumulative Weibull failure probability by log(time) for each group. Lines that are approximately linear empirically indicate the appropriateness of using a Weibull model for further analysis. See Exponential, Weibull, and Lognormal Plots and Fits.
Weibull Fit
Shows or hides the Extreme-Value Parameter Estimates and Weibull Parameter Estimates tables. This option also adds a linear fit to the Weibull cumulative distribution function in the Weibull Plot. The two tables contain two popular forms of Weibull estimates (Figure 14.5). The α parameter is the 0.632 quantile of the failure-time distribution. The extreme-value table shows a different parameterization of the same fit, where λ = ln(α) and δ = 1/β. See Exponential, Weibull, and Lognormal Plots and Fits.
LogNormal Plot
Shows or hides the LogNormal Plot, which shows the cumulative lognormal failure probability by log(time) for each group. Lines that are approximately linear empirically indicate the appropriateness of using a lognormal model for further analysis. See Exponential, Weibull, and Lognormal Plots and Fits.
LogNormal Fit
Shows or hides the LogNormal Parameter Estimates table. This option also adds a linear fit to the lognormal cumulative distribution function in the LogNormal Plot. The μ and σ parameters correspond to the mean and standard deviation of a normally distributed natural logarithm of the time variable. See Exponential, Weibull, and Lognormal Plots and Fits.
Fitted Distribution Plots
Shows or hides a set of plots for each fitted distribution. The set of plots includes the fitted survival function, the fitted density function, and the fitted hazard function. If you have not performed a fit, no plot appears. See Fitted Distribution Plots.
Competing Causes
Performs an estimation of the Weibull model using the specified causes to indicate a failure event and other causes to indicate censored observations. The fitted distribution appears as a dashed line in the Survival Plot. See Competing Causes.
Estimate Survival Probability
Estimates survival probabilities and confidence intervals for the specified time values using the fitted distributions.
Estimate Time Quantile
Estimates a time quantile and confidence intervals for each specified survival probability using the fitted distributions.
Save Estimates
Creates a new data table that contains survival and failure estimates, confidence intervals, and other distribution statistics for each group.