Enter your data for Regression with Life Data

Stat > Reliability/Survival > Regression with Life Data

In This Topic

Enter your data
Model terms
Assumed distribution

Enter your data

Select the option that best describes your data.

Responses are uncens/right censored data

If you have exact failure times, right-censored observations (you know only the time after which the failure occurred), or both, complete the following steps. For more information on censored data, go to Data censoring.

In Variables/Start variables, enter the column that contains the failure times or right censoring times. You can enter up to 10 columns (for 10 different samples).
If you have frequency data for each variable, in Freq. columns (optional), enter a column that indicates the number of units for each failure time or censoring time.

In this worksheet, the Time column contains the failure times. The Censor column contains the censoring indicators: an F designates an actual failure time; a C designates a unit that was removed from the test, and thus censored. The Temperature column contains the values for the predictor variable.

C1	C2	C3
Time	Censor	Temperature
53	F	125
60	F	125
53	F	125
99	C	125
51	F	150
40	F	150

Responses are uncens/arbitrarily censored data

If your data include left-censored observations (you know only the time before which the failure occurred), interval-censored observations (you know only the times between which the failure occurred), or a varied censoring scheme that includes exact failure times, right censoring, left censoring, and interval censoring, complete the following steps. For more information on censored data, go to Data censoring.

In Variables/Start variables, enter the column that contains the start times. You can enter up to 10 columns (for 10 different samples). The start times in the column depend on how the data are censored.

Observation	Value in Start column
Exact failure time	Failure time
Right censored	Time after which the failure occurred
Left censored	* (missing value symbol)
Interval censored	Time at start of interval during which the failure occurred

In End variables, enter the column that contains the end times. You can enter up to 10 columns (for 10 different samples). The end times in the column depend on how the data are censored.

Observation	Value in End column
Exact failure time	Failure time
Right censored	* (missing value symbol)
Left censored	Time before which the failure occurred
Interval censored	Time at end of interval during which the failure occurred

If you have frequency data for each variable, in Freq. columns (optional), enter a column that indicates the number of units for each failure time or censoring time.

In this worksheet, the Start column contains the start times and the End column contains the end times. The Temperature column contains the values for the predictor variable. The Frequency column (optional) indicates the number of units that are included in each interval. For example, at a temperature of 125°, 20 units are left censored at 10,000 hours. 2 units are exact failures at 30,000 hours. 26 units are interval censored between 30,000 and 40,000 hours. 190 units are right censored at 50,000 hours.

C1	C2	C3	C5
Start	End	Frequency	Temperature
*	10000	20	125
10000	20000	10	125
20000	30000	10	125
30000	30000	2	125
30000	40000	26	125
40000	50000	42	125
50000	*	190	125
*	10000	22	150
10000	20000	14	150
20000	30000	15	150
30000	40000	33	150
40000	50000	55	150
50000	*	161	150

Model terms

Model: Enter the terms that may explain or predict changes in the failure time of a product. The model can include categorical factors and covariates. The factors and covariates can be crossed or nested. For more information on specifying the model, go to Specifying the model terms in regression with life data.
Important
Minitab assumes that any variable in the model is a covariate unless the variable is specified as a factor. Therefore, if any predictors are factors, you must enter them again in Factors (optional).
Factors (optional): If any variables in the model are factors, you must enter them here. Enter a column for each factor in the model. Factors contain group classifications that may explain or predict changes in the response, and may be crossed or nested.
Note
Life data models in Minitab must be full rank and hierarchical. For more information, go to Restrictions on models for regression with life data.

Assumed distribution

Select a distribution to model your data. Base your decision on process knowledge or use probability plots to evaluate the model fit. For more information, go to Distribution fit for reliability analysis.