Data snooping in deformation surveys

The use of least squares estimation methods to provide accurate and reliable results in the analysis of deformation monitoring networks requires observational data that is random in nature. This necessities the elimination of all blunders to ensure that the data follows a normal distribution. The elimination of gross errors from observations is a complex problem, which may be solved in various ways i.e . by application of observational methods designed to prevent the occurrence of blunders; by safeguarding against erroneous data transfer; by testing the functional conditions which the observations in question must satisfy (eg. the sum of angles in a triangle must be 180 degrees); and finally by detection of blunders through statistical analysis of post-adjustments. A brief overview of the reliability in network design is given before investigating the statistic of residuals and method of data snooping.

PDF Data snooping in deformation surveys.pdf Restricted to Repository staff only Download (234kB)

Actions (login required)

View Item