Bride and Prejudice

As most of you probably don't know, a little movie staring Aishwarya Rai, opened in the US this past weekend. I saw a few weeks ago, and I didn't think it was great, i mean it wasn't bad, but it wasn't really awesome either. I guess it was an okay film.

Well the point really is, that for the first weekend, it ranked 20, with $385, 848*, far behind the number one film, Hitch which had $43,142,214* in sales. So I guess your thinking whats the point. I think the point is, even though most of you haven't heard of this film and it ranked 20th on the weekend gross, it did much better than what you think at first glance. Being banking you just start to think about everything in terms of ratios. Granted this isn't EV/EBITDA, but if you take Weekend Gross / # number of theaters, you can get a much better picture of what really is going on. I mean after you normalize weekend gross by number of theaters, you see Hitched came in at $12,067.8 per theater and Bride and Prejudice came in at $12,057.8 per theater. (3575* theaters for Hitched and 32* for Bride and Prejudice)

(*source : http://movies.yahoo.com/boxoffice/latest/rank.html)

Anyways, there is exactly a $10 difference. But I will now proceed to prove that this difference is statistically insignificant. We will have to implementing a two dementional t-test to test the significance between the two means.

t=[(x_1 bar - x_2 bar) - (Mu_1 - MU_2) ]/ [s_p^2 / n_1 + s_p^2/n_2]
where s_p^2 =[ (n_1 - 1)* s_1^2 + (n_2 - 1)* s_2^2]/(n_1 + n_2 -2)

Since we don't know the variance of the sample, lets just hold that constant as a variable and will address it in a bit. Let us also assume that the variances are the same for both samples. I think in the context of this framework it is realistic to assume this.

s_p^2 = 3605 S /3605 = S.
t =[ (12,067.8-12,057.8) - (0) ]/ ( .17757 * s )
t = 10/.17757*s
t=56.317/s

degrees of freedom = 3575 + 32 -2 = 3605. We can now look this all up in a t table.
Since we don't know the variance, we can back in to it depending on the significance level.
At the
.1 significance t = 1.282,
.05 significance t = 1.645,
.025 significance t = 1.96,
.01 significance t=2.326,
.005 significance t = 2.576.

So lets just test at the .005 level, to see what the implied standard deviation.
2.576= 56.317/s ; s = 21.86
2.326=56.317/s; s = 24.21
1.96 =56.317/s; s= 28.73

So clearly based on this we see that even at the .005 significance level, the standard deviation would have to be only 21.86. Which is completely reasonable. Meaning that there is a .5% chance we can reject the null hypothesis which is the means are the same.

So based on this we can clearly determine that Bride and Prejudice and Hitched did just as well.

So what was the point of this whole exercise, well to show that Bride Prejudice did well, that and Aishwarya Rai, is hot. Well i guess i didn't really prove that, but hell, thats the only really thing that matters