Description
When excludeIntercept is true, remove instrumentNames that occur literally in interceptNames; otherwise select all names.
Instrumental-variable and moment calculations compare supplied regressors, instruments, residuals and first-stage quantities. A local moment or relevance statistic does not establish exclusion, exogeneity, identification or a complete estimator.
When excludeIntercept is true, remove instrumentNames that occur literally in interceptNames; otherwise select all names.
When to use this tool
Use Instrument Intercept Exclusion Flag when checking one supplied first-stage or moment contribution.
How to use the tool and how it works
- Enter instrumentNames, interceptNames, excludeIntercept. Keep parallel arrays aligned. Use the units and conventions below.
- Run the tool. Correct any format, range, or cross-field input error before interpreting the output.
- Read selected, excluded. Use the output convention below.
When excludeIntercept is true, remove instrumentNames that occur literally in interceptNames; otherwise select all names. Return selected and excluded arrays in source order.1
| Input field | Meaning and units | Accepted input |
|---|---|---|
| instrumentNames | Instrument Names | Required. Array of text; maximum UTF-16 length 8192; maximum items 256. |
| interceptNames | Intercept Names | Required. Array of text; maximum UTF-16 length 8192; maximum items 256. |
| excludeIntercept | Exclude Intercept | Required. Boolean. |
| Output field | Interpretation | Returned shape |
|---|---|---|
| selected | Selected | Required. Array of text; maximum UTF-16 length 8192; maximum items 256. |
| excluded | Excluded | Required. Array of text; maximum UTF-16 length 8192; maximum items 256. |
{
"input": {
"instrumentNames": [
"const",
"z1",
"z2"
],
"interceptNames": [
"const"
],
"excludeIntercept": true
},
"output": {
"selected": [
"z1",
"z2"
],
"excluded": [
"const"
]
}
}Limitations and assumptions
- Duplicates are preserved. Names do not reveal actual constant columns, rank or validity, so this is a supplied label filter rather than design-matrix inspection.
- 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.
Alternative or Complementary methods
Fit an appropriate full model and examine identifying assumptions and diagnostics before interpreting an IV result.
References
Similar or alternative tools
- Instrument Relevance One Moment Gap
covariance = meanProduct-instrumentMean*regressorMean; gap = covariance-referenceCovariance.
- Weak Instrument First Stage Partial Ratio
Calculate nested first-stage partial R-squared and an ordinary homoskedastic partial F statistic from supplied sums of squares.