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 column in the input (or in the auxiliary) data set 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 auxiliary data set is present, then the number of rows of the input data set (X) must be equal to 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 in Y are removed.
7. The function RMNA preserves the original orders (rows and columns) of the data set (X).
8. Using the RMNA function with time-series data set (univariate or multivariate) is problematic, as output data set may not be equally 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.

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.