Description
observed clamps latent to [lower,upper]; leftCensored is latent <= lower and rightCensored is latent >= upper.
A censored response records a boundary rather than an unrestricted latent value outside that boundary. Clipping a known hypothetical value can illustrate the observation rule without estimating the hidden values in actual data.
observed clamps latent to [lower,upper]; leftCensored is latent <= lower and rightCensored is latent >= upper.
When to use this tool
Use Censored Outcome Indicator One Row when applying an explicit two-sided censoring convention to a supplied latent value.
How to use the tool and how it works
- Enter latent, lower, upper. Use the units and conventions below.
- Run the tool. Correct any format, range, or cross-field input error before interpreting the output.
- Read observed, leftCensored, rightCensored. Inspect leftCensored, rightCensored before interpreting the associated values.
Require upper > lower. observed clamps latent to [lower,upper]; leftCensored is latent <= lower and rightCensored is latent >= upper.1
| Input field | Meaning and units | Accepted input |
|---|---|---|
| latent | Latent | Required. Finite number. |
| lower | Lower bound under the stated coordinate, derivative or uncertainty convention. | Required. Finite number. |
| upper | Upper bound under the stated coordinate, derivative or uncertainty convention. | Required. Finite number. |
| Output field | Interpretation | Returned shape |
|---|---|---|
| observed | Position-aligned flags specifying which records are observed. | Required. Finite number. |
| leftCensored | True at or below lower, including equality. | Required. Boolean. |
| rightCensored | True at or above upper, including equality. | Required. Boolean. |
{
"input": {
"latent": -2,
"lower": 0,
"upper": 10
},
"output": {
"observed": 0,
"leftCensored": true,
"rightCensored": false
}
}Limitations and assumptions
- Equality is flagged as censored even without numerical change. This clips a supplied latent value, not identification of censoring from observed records or a fitted Tobit model.
- Inputs must satisfy both the field types and the cross-field conditions above. Invalid input causes an error; a successful false flag or documented placeholder is a different result.
- Numeric calculations use binary floating point unless exact decimal-integer arithmetic is explicitly stated. No hidden measurement tolerance or decimal rounding is added; nonfinite or out-of-contract results are rejected. Numeric signed-zero results are exported as zero.
Alternative or Complementary methods
Use a suitable censored-outcome model for estimation and distinguish actual censoring metadata from hypothetical clipping.
References
Similar or alternative tools
- Learner Response Time Censored Flag
recordedLatency = min(latency,timeout); censored tests latency >= timeout when equalityCensored, otherwise latency > timeout.
- Two Stage Least Squares One Fitted Value Term
fittedEndogenous = intercept+sum(instrument*coefficient).