Comments on: Predictive Analysis http://thomaslarock.com/2009/05/predictive-analysis/ Sat, 19 May 2012 00:44:00 +0000 hourly 1 http://wordpress.org/?v=3.3.2 By: Brett Flippin http://thomaslarock.com/2009/05/predictive-analysis/comment-page-1/#comment-615 Brett Flippin Wed, 06 May 2009 19:15:05 +0000 http://sqlbatman.com/?p=1827#comment-615 Your comment cleared quite a bit up in my understanding. For me, predicting the future is saying that with 100% certainty something is going to occur. Predictive analysis is the act of finding your "best guess" or what you expect to happen, based on the analysis of data you've already gathered. You aren't predicting with 100% certainty unless your data is so good and complete that you can. The likelihood of which is virtually impossible. I guess that's where I differ because I wouldn't call predictive analysis predicting the future, it's more of this is what I think is going to happen in the future with a reasonable margin of error. I agree absolutely that people that wholly trust their mining models are missing the point of predictive analysis entirely. Analysis can give you insight, but only an intelligent person with a good mind for data can make correct decisions given all the data and all the analysis in the world. Your comment cleared quite a bit up in my understanding. For me, predicting the future is saying that with 100% certainty something is going to occur. Predictive analysis is the act of finding your “best guess” or what you expect to happen, based on the analysis of data you’ve already gathered. You aren’t predicting with 100% certainty unless your data is so good and complete that you can. The likelihood of which is virtually impossible.

I guess that’s where I differ because I wouldn’t call predictive analysis predicting the future, it’s more of this is what I think is going to happen in the future with a reasonable margin of error.

I agree absolutely that people that wholly trust their mining models are missing the point of predictive analysis entirely. Analysis can give you insight, but only an intelligent person with a good mind for data can make correct decisions given all the data and all the analysis in the world.

]]> By: SQLBatman http://thomaslarock.com/2009/05/predictive-analysis/comment-page-1/#comment-614 SQLBatman Wed, 06 May 2009 15:21:07 +0000 http://sqlbatman.com/?p=1827#comment-614 Brett, The point of my post was to hope that someone like yourself would read it and help me to understand a little more about the use of statisical analysis. drop me an email as i would enjoy discussing this with you further. when you said "The idea isn’t to predict the future, but to give you an idea of what to expect.", how is that different than predicting the future? If I said that we could expect there to be a tornado in a trailer park next week, isn't that also predicting the future to some degree? I understand the idea behind CI's, and if there is one point i would like to make it is that you are not going to ever have a perfect system, a perfect model, and you should not be trusting people to make business decisions that end up putting all of their eggs in one basket as a result of predicitve analysis. Brett,

The point of my post was to hope that someone like yourself would read it and help me to understand a little more about the use of statisical analysis. drop me an email as i would enjoy discussing this with you further.

when you said “The idea isn’t to predict the future, but to give you an idea of what to expect.”, how is that different than predicting the future? If I said that we could expect there to be a tornado in a trailer park next week, isn’t that also predicting the future to some degree?

I understand the idea behind CI’s, and if there is one point i would like to make it is that you are not going to ever have a perfect system, a perfect model, and you should not be trusting people to make business decisions that end up putting all of their eggs in one basket as a result of predicitve analysis.

]]> By: Brett Flippin http://thomaslarock.com/2009/05/predictive-analysis/comment-page-1/#comment-613 Brett Flippin Wed, 06 May 2009 13:22:43 +0000 http://sqlbatman.com/?p=1827#comment-613 It seems that you might have missed the point of predictive analytics. The idea isn't to predict the future, but to give you an idea of what to expect. It's not meant to be taken with absolute authority. That's why we have things like confidence intervals and that's why you test and re-test your mining models to see if they are producing results in the range that you deem accurate. Predictive analysis is based on taking the data that you have, and trying to extrapolate what that might mean for you in the future using that data. Statistical analysis can be used in predictive analysis and should be. You have to understand what has already happened to understand what might happen the future. Amazon is taking into account your past searches, wish list items, past purchases, etc, and trying to predict what you might want to buy next and putting in front of you. They do this by analyzing all the data they've gathered on you in the past and trying to predict what you might want now (as a future sale for them). I'm not sure I 100% understand the point of your post, so I could be entirely off base, you might be thinking about it too much. It seems that you might have missed the point of predictive analytics. The idea isn’t to predict the future, but to give you an idea of what to expect. It’s not meant to be taken with absolute authority. That’s why we have things like confidence intervals and that’s why you test and re-test your mining models to see if they are producing results in the range that you deem accurate.

Predictive analysis is based on taking the data that you have, and trying to extrapolate what that might mean for you in the future using that data. Statistical analysis can be used in predictive analysis and should be. You have to understand what has already happened to understand what might happen the future. Amazon is taking into account your past searches, wish list items, past purchases, etc, and trying to predict what you might want to buy next and putting in front of you. They do this by analyzing all the data they’ve gathered on you in the past and trying to predict what you might want now (as a future sale for them).

I’m not sure I 100% understand the point of your post, so I could be entirely off base, you might be thinking about it too much.

]]> By: Brent Ozar http://thomaslarock.com/2009/05/predictive-analysis/comment-page-1/#comment-612 Brent Ozar Tue, 05 May 2009 17:47:43 +0000 http://sqlbatman.com/?p=1827#comment-612 I wonder why they don't put the Tylenol next to the condoms, because every time I want to - wait - never mind. I wonder why they don’t put the Tylenol next to the condoms, because every time I want to – wait – never mind.

]]> By: SQLBatman http://thomaslarock.com/2009/05/predictive-analysis/comment-page-1/#comment-611 SQLBatman Tue, 05 May 2009 16:37:29 +0000 http://sqlbatman.com/?p=1827#comment-611 perhaps a bacon and basil sandwich? perhaps a bacon and basil sandwich?

]]> By: wnylibrarian http://thomaslarock.com/2009/05/predictive-analysis/comment-page-1/#comment-610 wnylibrarian Tue, 05 May 2009 16:33:39 +0000 http://sqlbatman.com/?p=1827#comment-610 Actually, for the record, I said "Produce" for the basil question. It's "sooooo" the librarian in me to categorize, I realize that. That said so many people I've found use software and devices not in the way they were originally intended. Predictive Analysis? Seems next to impossible to me. There are always variables that one cannot possibly think of. Now, time for lunch. I smell some bacon cooking :-) Actually, for the record, I said “Produce” for the basil question. It’s “sooooo” the librarian in me to categorize, I realize that. That said so many people I’ve found use software and devices not in the way they were originally intended. Predictive Analysis? Seems next to impossible to me. There are always variables that one cannot possibly think of. Now, time for lunch. I smell some bacon cooking :-)

]]> By: SQLBatman http://thomaslarock.com/2009/05/predictive-analysis/comment-page-1/#comment-609 SQLBatman Tue, 05 May 2009 15:31:32 +0000 http://sqlbatman.com/?p=1827#comment-609 i like the example of the correlation made between diaper sales and beer sales, spiking on friday nights as new fathers are told to stop and pick up a few things on their way home from work. i like the example of the correlation made between diaper sales and beer sales, spiking on friday nights as new fathers are told to stop and pick up a few things on their way home from work.

]]> By: Aaron Alton http://thomaslarock.com/2009/05/predictive-analysis/comment-page-1/#comment-608 Aaron Alton Tue, 05 May 2009 15:20:26 +0000 http://sqlbatman.com/?p=1827#comment-608 People who use data mining in retail product placement piss me off. It's a good idea online, because when you search for "basil", you'll still find the damn basil. But when they put it beside the tomatoes, even if you're looking for tomatoes and basil, you're liable to walk right by it. What's next - beano in the chili aisle? Migrane medication with the baby food? Ex-Lax with the bacon? People who use data mining in retail product placement piss me off. It’s a good idea online, because when you search for “basil”, you’ll still find the damn basil. But when they put it beside the tomatoes, even if you’re looking for tomatoes and basil, you’re liable to walk right by it.

What’s next – beano in the chili aisle? Migrane medication with the baby food? Ex-Lax with the bacon?

]]>