Analytic techniques Flashcards
What’s the difference between Factor Analysis and PCA?
While both FA and PCA reduce data complexity, their goals differ. PCA aims to capture the most data variance in fewer dimensions. In contrast, FA assumes unobserved factors influence observed variables and aims to identify and measure these factors giving them meaningful interpretations. Think of PCA as summarizing data efficiently, while FA tries to uncover hidden influences.
What is a factor?
A linear combination of the original variables. Factors also represent the underlying dimension’s (constructs) that summarize or account for the original set of variables.
What are the outcomes of factor analysis?
Data summarization and data reduction. In summarizing the data, FA describes the data in a much smaller numbers of concepts than the original variables. Data reduction extends this process by deriving a factor score for each dimension (factor) and then substituting this value for the original values.
Are the dependent and independent variables in factor analysis?
No. In FA, all variables are simultaneously considered with no distinction to dependence or independence (e.g., predictor(s) vs. dependent variables). FA does not predict, it seeks to maximize the explanation of the entire dataset.
What is R squared?
It’s the percentage of variation in Y that is explained By the variation in X. It’s a number between 0 and 1.
What is adjusted R squared?
It’s R squared adjusted for the number of x variables. If you are comparing models that have multiple X variables, It’s better to use adjusted R square. Adjust the calculation of R squared based on the number of variables in the model. Otherwise adding more variables usually makes R squared go up, which can be misleading
P-value
How can we be confident that the regression coefficients are statistically significant? For that we look at the P value. Each coefficient that we’ve estimated whether it’s the intercept or the two slope coefficients, each of them has its own P value. To remind you we’re looking for P values which are less than 5% less than 0.05
F-test
The F-test is primarily used to assess if your regression model as a whole is a good fit for the data. It tests whether at least one of your independent variables has a significant relationship with the dependent variable. It tells you if the x variables are jointly significant.
Working capital
Current assets - current liabilities
Cash conversion cycle
DSO+DIO-DPO
Change in working capital
Increase in current assets = deduction to operating cash flow (OCF)
Increase in current liabilities = increase to OCF
Operating cash flow margin
Measures how much cash is generated for each dollar of revenue.
Formula: Operating cash flow/Net revenue
