- This topic has 28 replies, 6 voices, and was last updated 16 years, 10 months ago by
drunkle.
-
AuthorPosts
-
November 13, 2007 at 4:26 PM #10897November 13, 2007 at 5:41 PM #99145
drunkle
Participanti want to know what’s with the weird breakdown:
Middle 1/3 (482K to 678K)
like, there are homes that broke through the low and up into the middle, but will fall back down. likewise, the plethora of middle homes that are deep into high territory yet.
i’m assuming that the breaks are according to the actual number of homes vs price, but… matter of fact, how they did their analysis at all is confusing. how do you track price changes when price is part of your categorization?
maybe if they had made the spreads strictly by price spread regardless of number of homes at a particular price point, would be easier to visualize…
November 13, 2007 at 5:41 PM #99207drunkle
Participanti want to know what’s with the weird breakdown:
Middle 1/3 (482K to 678K)
like, there are homes that broke through the low and up into the middle, but will fall back down. likewise, the plethora of middle homes that are deep into high territory yet.
i’m assuming that the breaks are according to the actual number of homes vs price, but… matter of fact, how they did their analysis at all is confusing. how do you track price changes when price is part of your categorization?
maybe if they had made the spreads strictly by price spread regardless of number of homes at a particular price point, would be easier to visualize…
November 13, 2007 at 5:41 PM #99222drunkle
Participanti want to know what’s with the weird breakdown:
Middle 1/3 (482K to 678K)
like, there are homes that broke through the low and up into the middle, but will fall back down. likewise, the plethora of middle homes that are deep into high territory yet.
i’m assuming that the breaks are according to the actual number of homes vs price, but… matter of fact, how they did their analysis at all is confusing. how do you track price changes when price is part of your categorization?
maybe if they had made the spreads strictly by price spread regardless of number of homes at a particular price point, would be easier to visualize…
November 13, 2007 at 5:41 PM #99229drunkle
Participanti want to know what’s with the weird breakdown:
Middle 1/3 (482K to 678K)
like, there are homes that broke through the low and up into the middle, but will fall back down. likewise, the plethora of middle homes that are deep into high territory yet.
i’m assuming that the breaks are according to the actual number of homes vs price, but… matter of fact, how they did their analysis at all is confusing. how do you track price changes when price is part of your categorization?
maybe if they had made the spreads strictly by price spread regardless of number of homes at a particular price point, would be easier to visualize…
November 13, 2007 at 7:35 PM #99165greensd
ParticipantAnd remember, the percentages will be a lot smaller on the way down. The lower tier went up 200%, but would only have to go down 66% to get back where it started, ignoring inflation. If you take inflation into account, we’re even closer to 2001 prices.
Just tracking the segment boundaries could be interesting. When the bottom of the middle third hits $400,000, maybe I’ll start looking around.
November 13, 2007 at 7:35 PM #99227greensd
ParticipantAnd remember, the percentages will be a lot smaller on the way down. The lower tier went up 200%, but would only have to go down 66% to get back where it started, ignoring inflation. If you take inflation into account, we’re even closer to 2001 prices.
Just tracking the segment boundaries could be interesting. When the bottom of the middle third hits $400,000, maybe I’ll start looking around.
November 13, 2007 at 7:35 PM #99248greensd
ParticipantAnd remember, the percentages will be a lot smaller on the way down. The lower tier went up 200%, but would only have to go down 66% to get back where it started, ignoring inflation. If you take inflation into account, we’re even closer to 2001 prices.
Just tracking the segment boundaries could be interesting. When the bottom of the middle third hits $400,000, maybe I’ll start looking around.
November 13, 2007 at 7:35 PM #99242greensd
ParticipantAnd remember, the percentages will be a lot smaller on the way down. The lower tier went up 200%, but would only have to go down 66% to get back where it started, ignoring inflation. If you take inflation into account, we’re even closer to 2001 prices.
Just tracking the segment boundaries could be interesting. When the bottom of the middle third hits $400,000, maybe I’ll start looking around.
November 13, 2007 at 8:48 PM #99259Rich Toscano
KeymasterDrunkle – The price breakpoints are as of August 2007. It means that a given home’s category was determined as of that date (either by the sale price if it sold in Aug 2007, or the imputed price based on the HPI), and it will never jump categories regardless of where its price goes.
The problem you describe, of a home jumping categories, would happen if they were measuring median price or something. But they can avoid it because they are comparing same-home sales, so they can “pin” a given home into one category even if the price changes.
In retrospect, I’ll bet there was a much better way to explain that. π
Rich
PS – As for the weird breakpoints, I assume they are just using a distribution of some sort so that there are an equal number of homes in the hi and low categories, or something.
PPS – All of the above is speculation on my part. But I’ll bet it’s true.
November 13, 2007 at 8:48 PM #99265Rich Toscano
KeymasterDrunkle – The price breakpoints are as of August 2007. It means that a given home’s category was determined as of that date (either by the sale price if it sold in Aug 2007, or the imputed price based on the HPI), and it will never jump categories regardless of where its price goes.
The problem you describe, of a home jumping categories, would happen if they were measuring median price or something. But they can avoid it because they are comparing same-home sales, so they can “pin” a given home into one category even if the price changes.
In retrospect, I’ll bet there was a much better way to explain that. π
Rich
PS – As for the weird breakpoints, I assume they are just using a distribution of some sort so that there are an equal number of homes in the hi and low categories, or something.
PPS – All of the above is speculation on my part. But I’ll bet it’s true.
November 13, 2007 at 8:48 PM #99243Rich Toscano
KeymasterDrunkle – The price breakpoints are as of August 2007. It means that a given home’s category was determined as of that date (either by the sale price if it sold in Aug 2007, or the imputed price based on the HPI), and it will never jump categories regardless of where its price goes.
The problem you describe, of a home jumping categories, would happen if they were measuring median price or something. But they can avoid it because they are comparing same-home sales, so they can “pin” a given home into one category even if the price changes.
In retrospect, I’ll bet there was a much better way to explain that. π
Rich
PS – As for the weird breakpoints, I assume they are just using a distribution of some sort so that there are an equal number of homes in the hi and low categories, or something.
PPS – All of the above is speculation on my part. But I’ll bet it’s true.
November 13, 2007 at 8:48 PM #99183Rich Toscano
KeymasterDrunkle – The price breakpoints are as of August 2007. It means that a given home’s category was determined as of that date (either by the sale price if it sold in Aug 2007, or the imputed price based on the HPI), and it will never jump categories regardless of where its price goes.
The problem you describe, of a home jumping categories, would happen if they were measuring median price or something. But they can avoid it because they are comparing same-home sales, so they can “pin” a given home into one category even if the price changes.
In retrospect, I’ll bet there was a much better way to explain that. π
Rich
PS – As for the weird breakpoints, I assume they are just using a distribution of some sort so that there are an equal number of homes in the hi and low categories, or something.
PPS – All of the above is speculation on my part. But I’ll bet it’s true.
November 14, 2007 at 6:20 AM #99232raptorduck
ParticipantThe index uses sales pairs, two actual sales of the same home over time.
November 14, 2007 at 6:20 AM #99295raptorduck
ParticipantThe index uses sales pairs, two actual sales of the same home over time.
-
AuthorPosts
- You must be logged in to reply to this topic.