QUESTION 5

Fisheries management, especially where pelagic species are concerned is a vexing issue. The way commercial whaling destroyed whale stocks is well documented. The plight of our tuna fisheries is generally less well known. The data file tuna2 contains information on the tuna catch per unit of fishing effort through time. The 'St.Effort' variable has been calculated by taking information on the total tuna catch and dividing through by a measure of the size of the tuna fishing fleet and the time the fleet is at sea. This variable is then a standardised measure of the catch. When you plot the data you will find that there is more than one plausible relationship. For this question the model form you are to use is written as:

log(Yt)=+1Tt+et, whereYtis the standardised catch andTtis time. In this model the1estimate multiplied by 100 has an interpretation as an approximate percentage change. Autocorrelation is almost always a problem with time series data. Conduct a test to establish whether or not autocorrelation is an issue.

Using the robust SE approach to dealing with autocorrelation, what is the estimate of the slope? You need to report the value to three decimal places; do not use scientific notation (This is quiz question 10) (0.25 marks).

For the robust SE approach to dealing with autocorrelation, to one decimal place, what is the t-statistic value for the test where the null hypothesis is that the slope is zero? (This is quiz question 11) (0.25 marks).

Using the gls() approach for dealing with autocorrelation, what is the estimate of the slope? You need to report the value to three decimal places; do not use scientific notation (This is quiz question 12) (0.25 marks).

For the gls() approach to dealing with autocorrelation, to one decimal place, what is the t-statistic value for the test where the null hypothesis is that the slope is zero? (This is quiz question 13) (0.25 marks).

show using R