RMNA - Remove Missing Values

Returns an array of cells of a time series after removing all missing values.

Syntax

RMNA(X, Y)

X

is the input data sample (one- or two-dimensional array of cells).

Y

is an auxiliary (optional) data sample (one- or two-dimensional array of cells).

Remarks

1. Each input (or auxiliary) data set column corresponds to a separate variable.
2. Each row in the input (or in the auxiliary) data set corresponds to an observation.
3. The auxiliary data set (Y) is optional and may include one or more variables (columns).
4. The auxiliary data set input is required when both X and Y define observations. For example, consider a multiple regression data set where X corresponds to the explanatory variables, and Y is the response variable.
5. When an auxiliary data set is present, the number of rows of the input data set (X) must equal the number of rows of the auxiliary data set (Y).
6. Observations (i.e., rows) with missing values (e.g., #N/A, #VALUE!, #NUM!, empty cell) in either X or Y are removed.
7. The function RMNA preserves the data set's original orders (rows and columns) (X).
8. Using the RMNA function with a time-series (univariate or multivariate) data set is problematic, as the output data set may be unevenly spaced (over time).
9. For the time series-based data set, it is recommended to use a time index (e.g., timestamp) as one of the variables (columns) in the input data set.
10. The function RMNA has been revised in NumXL MARTHA version 1.67.

Files Examples

Related Links

• NumXL Tips & Tricks Blog - Removing Missing Values (RMNA).
• What to do About Missing Values in Time Series Cross-Section Data, James Honaker and Gary King, American Journal of Political Science.

References

• Hamilton, J.D.; Time Series Analysis, Princeton University Press (1994), ISBN 0-691-04289-6.
• Kenney, J. F. and Keeping, E. S. (1962) "Linear Regression and Correlation." Ch. 15 in Mathematics of Statistics, Pt. 1, 3rd ed. Princeton, NJ: Van Nostrand, pp. 252-285.