NAME: REMOVE_OUTLIERS_REPAIR PURPOSE: Routine repairs outliers. Quadratic trend is determined in a hollow vicinity of each point. The data value is compared with the trend value. If the deviation is statistically improbable, the value is repaired. There are 6 options for repair. CATEGORY: Data Processing CALLING SEQUENCE: repair, valneib, tneib, valiin, nmax, valiout INPUTS: VALNEIB: array of the data values in the hollow vicinity of the point. TNEIB: array of the observation times for the above values. VALIIN: the value to filter. NMAX: maximal probable deviation from the average in units of standard deviation KEYWORDS: None PARAMETERS: The code has one parameter "sch" setting the way outlier is repaired. OUTPUTS: VALIOUT: filtered value. DEPENDENCIES: None. Called by remove_outliers.pro MODIFICATION HISTORY: Written by: Vladimir Kondratovich 2007/12/28.