Possible effects of missing data on the statistics of wave climate are considered here in this paper. Gaps between recorded wave heights are filled in using estimates either from interpolation, or through ARMA modelling. Statistical models found in the literature are then used to study the possible fits of both the original and the full-length records. Preliminary results have shown that the amount of the missing data seems to have only limited effects on the statistics of the wave heights. However, since only one data set is used for the study, the present conclusion is inconclusive.
When handling with measurements, one has to deal with the problem of missing data as a rule. Data missing can occur due to a lot of reasons. In social science, when a survey is conducted, the person who is filling the questionnaire may refuse to answer some of the questions. This non-response will then cause gaps. On the other hand, in environmental studies, there are several reasons that can cause the missingness of the data. The measuring instruments, for example, may be malfunctioned; or the weather condition was too severe to conduct measurements. It can even be the case that some data in a record were entered erroneously and have to be deleted. In all the possible causes mentioned above, the invaluable data are lost forever. Depending on the circumstances of the missing data, the effects on the subsequent analyses may or may not be serious. For example, when the amount of data missing in a file is not large, the effect may not be severe. However, when a substantial amount of data is missing, the evaluated results may then be biased. This may not be the case for extraordinary happenings where data may become vital for the researchers.