Characteristic |
Beta |
95% CI 1 |
|---|---|---|
| (Intercept) | 880 | -1,167, 2,936 |
| genre | ||
| genrefiction | 1,072 | -988, 3,122 |
| genrenonfiction | 435 | -1,738, 2,604 |
| 1
CI = Credible Interval |
||
Model
I used a Gaussian Causal Model to calculate how genre, author rating, and published language relate to book sales. The model is based off this formula:
\[ y_i = \beta_{0} + \beta_{1} (chosen\ factor)_i + \epsilon_i \] Details about the model used—
The first graph—
The second graph—
Characteristic |
Beta |
95% CI 1 |
|---|---|---|
| (Intercept) | 749 | 361, 1,154 |
| author_rating | ||
| author_ratingFamous | 379 | -732, 1,490 |
| author_ratingIntermediate | 1,684 | 1,198, 2,171 |
| author_ratingNovice | 3,044 | 1,616, 4,500 |
| 1
CI = Credible Interval |
||
The third graph—
Characteristic |
Beta |
95% CI 1 |
|---|---|---|
| (Intercept) | 1,039 | -28, 2,156 |
| language_code | ||
| language_codeara | 158 | -5,378, 5,878 |
| language_codeenMCA | 708 | -2,346, 3,838 |
| language_codeenMGB | 346 | -1,400, 2,128 |
| language_codeenMUS | 866 | -335, 2,033 |
| language_codeeng | 846 | -294, 1,972 |
| language_codefre | 222 | -3,869, 4,188 |
| language_codenl | -679 | -8,520, 7,152 |
| language_codespa | 3,948 | -1,857, 9,543 |
| 1
CI = Credible Interval |
||
(The three factors I picked were language_codeeng, language_codespa, and language_codefre, which correspond to English, Spanish, and French respectively.)