Regression estimation equation term. The sequence number of the estimation equation term. The agency code of the data descriptor site, which contains the data used in the term calculation. The site number of the data descriptor site, which contains the data used in the term calculation.. The data descriptor of the data descriptor site, which contains the data used in the term calculation. The code indicating the type of statistic the value or values represent. The Estimation equation term coefficient value. Estimation equation term number of lag days. The number of days an estimation equation is lagged. t-ratio. The ratio of the computed regression coefficient to its standard deviation. It is used to test the null hypothesis that the coefficient is zero. Variance Inflation Factor (VIF). A common measure of multicollinearity (redundancy in predictors). VIFj = 1/(1-Rj^2) where Rj^2 is the coefficient of determination from a regression of predictor Xj on the other predictors in the regression. It represents an "inflation factor" for the variance of the regression coefficient due to redundancy of the predictors. An equation used to estimate daily values from other daily values. The code indicating the statistic type of the data being estimate. The sequence number of the estimation equation for the given data descriptor and statistic code. The effective starting date and time of the estimation equation. The effective ending date and time of the estimation equation. The name of the estimation equation. The description of the estimation equation. Estimation equation season start day. Estimation equation season end day. A true or false flag indicating if a logarithmic transform is used for the estimation equation computation. Equation constant value. The value of a in a regression equation of the form: y(t) = a + a11*x1(t-L1) + a12*x1(t-L2) +... + a21*x2(t-L1) +... + ank*xn(t-Lk) where y(t) = flow at "study" station at time t, xi(t-Lj) = flow at index station #i at time t-Lj, aij = coefficients to be estimated, and Lj = lag #j (e.g., 3 days, or -1 day, or zero, etc.). The "coefficient of correlation" and expresses the closeness of the relationship. It is the fraction of the total variance of the Y values that is expressed by the regression line. The value of one (1) is a perfect relationship. Adjusted R-squared value. A rescaling of R-square by degrees of freedom in the regression. It involves the ratio of mean squares rather than sums of squares. It is more comparable than R-square for regression models involving different numbers of explanatory variables with the same dependent variable. Standard error of estimate (SEE). The standard deviation of the residuals and describes the dependability of the estimates.