Below is my answer: Logistic regression analysis was used to find the factors that have a substantial impact on the likelihood that a transaction is fraudulent. (Model-> Limited Dependent Variable-> Logit-> Binary) Initially, I converted the variables "value" and "numdayopen" to their logarithmic form in order to take into consideration any potential non-linear relationship they might have with the dependent variable "isfraud."

At the 5% significance level, the logistic regression analysis showed that the variables 'numdayopen,' 'log(value),' 'dayind,' and 'international' have a statistically significant impact on the likelihood that a transaction is fraudulent. These variables' coefficients were found to be 'log(numdayopen)' -1.41734, 'log(value)' -0.0311325, 'international' 2.26706, 'dayind' 0.249493, with p-values less than 0.05. This implies that these factors significantly affect the possibility that a transaction is fraudulent.

Dependent variable: isfraud Standard errors based on Hessian

coefficient std. error z slope ----------------------------------------------------------------- const 3.21335 0.603678 5.323 l_numdayopen 1.41734 0.113437 12.49 0.0661798 l_value 0.0311325 0.0889498 0.3500 0.00145367 international 2.26706 0.402372 5.634 0.240119 dayind 0.249493 0.0738322 3.379 0.0116496

Mean dependent var 0.193750 S.D. dependent var 0.395483 McFadden R-squared 0.588283 Adjusted R-squared 0.575570 Log-likelihood 161.9250 Akaike criterion 333.8501 Schwarz criterion 357.2731 Hannan-Quinn 342.8482

Number of cases 'correctly predicted' = 715 (89.4%) f(beta'x) at mean of independent vars = 0.047 Likelihood ratio test: Chi-square(4) = 462.734 [0.0000] Above is the data used.

p-value=2(1(z))

Using the formula below, we can find out the p-values for each variable.

The p-value for 'log(numdayopen)' is shown by the following working out:

According to the z-score for log(numdayopen), we can see that 100% of the distribution lies to the left of a z-score of 12.49 therefore the p-value is 0.

The p-value for 'log(value)' is shown by the following working out:

2(10.6368)=0.7264

The p-value for 'international' is shown by the following working out:

According to the z-score for 'international', we can see that 100% of the distribution lies to the left of a z-score of 6.634 therefore the p-value is 0.

The p-value for 'dayind' is shown by the following working out:

2(10.9993)=0.0014

This implies that to the 5% significance level, the variables 'dayind' and log(numdayopen) significantly affect the possibility that a transaction is fraudulent. The next writing will be the actual answer, tell me whether ive done everything right or not.

(a) Determining Variables Affecting Fraudulent Transactions: I used a logistic regression analysis to find the factors that have a substantial impact on the likelihood that a transaction is fraudulent. Initially, I converted the variables "value" and "numdayopen" to their logarithmic form in order to take into consideration any potential non-linear relationship they might have with the dependent variable "isfraud."

At the 5% significance level, the logistic regression analysis showed that the variables "numdayopen," "log(value)," "dayind," and "international" have a statistically significant impact on the likelihood that a transaction is fraudulent. These variables' coefficients were found to be [coefficients] with p-values less than 0.05. This implies that these factors significantly affect the possibility that a transaction is fraudulent.