Brand choice and logit models - Chi-square

7 important questions on Brand choice and logit models - Chi-square

What are cross-tabs and to which variables are they applicable?

  • The collection of the frequencies with which the ith level of the row variable occurs in a sample with the jth level of the column variable.
  • Suited to categorical variables or ordinal variables or categorized versions of continuous variables.

How do you denote the sums of rows, columns and total?

  • Row sums: ni+
  • Column sums: n+j
  • Total sum: n++

What is the probability that a customer is in cell (i,j)?

  • pij=pi+*p+j
  • pi+=ni+/n++
  • p+j=n+j/n++
  • pij=(ni+/n++)(n+j/n++)
  • Higher grades + faster learning
  • Never study anything twice
  • 100% sure, 100% understanding
Discover Study Smart

What is the X2 fit statistic?


With df=(I-1)(J-1).

What are residuals and standardized residuals?

Residuals are differences between our data (oij) and expected frequencies (eij).
Standardized residuals are (oij - eij)/sqr(eij). Is significant if it exceeds 1.96, meaning no independence.
  • High positive standardized residuals mean likely to buy again.
  • High negative standardized residuals mean unlikely to buy again.

How do you make the eij formula linear?

By taking the natural logarithm of both sides:
  • ln(eij) = ln(ni+) + ln(n+j) - ln(n++)

What are log linear models used for?

To analyze cross-tabs that have more than 2 dimensions.

The question on the page originate from the summary of the following study material:

  • A unique study and practice tool
  • Never study anything twice again
  • Get the grades you hope for
  • 100% sure, 100% understanding
Remember faster, study better. Scientifically proven.
Trustpilot Logo