Upgrading from:

Statistics`NonlinearFit`

NonlinearFit is replaced by FindFit.
NonlinearRegress functionality is now available using the built-in function NonlinearModelFit.

NonlinearFit is replaced by FindFit:

Version 5.2 << Statistics`NonlinearFit`; NonlinearFit[{{1, 3}, {2, 8}, {3, 4}, {4, 0}}, a*Sin[b*x], {x}, {a, b}]

The NonlinearRegress function from Statistics`NonlinearFit` has been superseded by NonlinearModelFit:

Version 5.2 << Statistics`NonlinearFit`; NonlinearRegress[{{1, 3}, {2, 8}, {3, 4}, {4, 0}}, a*Sin[b*x], {x}, {a, b}]

The arguments are consistent with those of FindFit:

The result is returned as a FittedModel object, of which properties can be returned:

For a list of properties available, use "Properties":

Version 5.2 << Statistics`NonlinearFit`; RegressionReportValues

To get the functional form of the FittedModel object, use Normal:

Some of the properties previously available to NonlinearRegress have been renamed:

NonlinearRegressNonlinearModelFit
AsymptoticCorrelationMatrix"CorrelationMatrix"
AsymptoticCovarianceMatrix"CovarianceMatrix"
MeanPredictionCITable"MeanPredictionConfidenceIntervalTable"
ParameterCITable"ParameterConfidenceIntervalTable"
SinglePredictionCITable"SinglePredictionConfidenceIntervalTable"
SummaryReport{"BestFitParameters", "ParameterConfidenceIntervalTable", "EstimatedVariance", "ANOVATable", "CorrelationMatrix", "FitCurvatureTable"}