0

# Value Statistical Significance

Sun 8 Feb, 2015 03:36 pm
Hi,

I am looking at some studies ahead of starting my own. The study relates to analysing market reactions and refining samples to understand whether subsamples can be associated with certain market reactions.

I determine whether the market reaction is statistically different from zero (using t-test and p-value). However, I am aware of some studies not only calculating if one reaction is significant, but whether the difference between two subsamples is significantly different from zero.

Is anyone on here able to explain how the difference is interpreted? And also in what cases would you choose to calculate if two subsamples are significantly different.

Hopefully this makes sense.

• Topic Stats
• Top Replies
Type: Question • Score: 0 • Views: 1,627 • Replies: 2
No top replies

Rickoshay75

0
Sun 8 Feb, 2015 04:25 pm
@ryanmc29,
ryanmc29 wrote:

Hi,

I am looking at some studies ahead of starting my own. The study relates to analysing market reactions with certain market reactions.

It's not that complicated, all you need to do is to chart a day by day active stock for a month, determine trend, up or down, buy it against the trend, limit your profits and losses, or buy stock options, up or down, against the trend after a two or three point movement either way, and sell it the next day of trading. remember, always go against the trend!! Look upon a trend as a stretched out rubber band building up reverse reaction.

“You want people walking away from the conversation with some kernel of wisdom or some kind of impact.” Henry Dean Stanton
ryanmc29

1
Wed 11 Feb, 2015 02:55 pm
@Rickoshay75,
Hi Rickoshay75,

Thanks for your response. Apologies- my post wasn't making the point I was wanting to.

Hopefully this may clear it up a bit.

A scenario:
- I have 2 samples (x and y). I find that x reacts significantly to a certain market event, but y is associated with a non- significant market reaction.
(this is investigating whether these are, individually, significantly different from zero.)

Then, I calculate the difference between x and y and find this to be non- significant.
Are you able to advise me how I would interpret this?

Thanks in advance to anyone who can contribute

R
0 Replies

### Related Topics

1 in 5 women get raped? - Discussion by maxdancona
How will you die? - Discussion by Robert Gentel
Statistics - probability - Question by jrl06
Sampling formula - Question by Moses mbeya

1. Forums
2. » Value Statistical Significance