mb
2 楼
帮你查了下目录，$0.4
Guatemala是中美不是南美
想知道价格找scott目录
【在 s**a 的大作中提到】
: 家里有一些南美邮票，不知道有没有价值，我贴些照片。
: 如果在美国要搜寻这些邮票的价格，有什么途径？
Guatemala是中美不是南美
想知道价格找scott目录
【在 s**a 的大作中提到】
: 家里有一些南美邮票，不知道有没有价值，我贴些照片。
: 如果在美国要搜寻这些邮票的价格，有什么途径？
t*y
3 楼
In my opinion, about 95% of the stamps in the world has almost no value (
below $5). About 4.5% has some value ($5-20), and the top 0.5% which has
value of $20-$10000+, accounts for 95% value of all stamps.
Check China/PRC stamps and this figure is generally applied.
With that, that's usually no need to check the values of most stamps,
especially those came in a album. I have a number of international albums
with 3000-5000 stamps, mint and used, from 1880-1970, total worth is less
than $50.
A f
below $5). About 4.5% has some value ($5-20), and the top 0.5% which has
value of $20-$10000+, accounts for 95% value of all stamps.
Check China/PRC stamps and this figure is generally applied.
With that, that's usually no need to check the values of most stamps,
especially those came in a album. I have a number of international albums
with 3000-5000 stamps, mint and used, from 1880-1970, total worth is less
than $50.
A f
z*9
4 楼
If you can sell those no value stamps, it is still very good because of its
volume..
volume..
h*s
5 楼
说得挺好，这个现象在很多领域都存在，比如城市的大小，单词使用频率，
被下载的文件等等，数学上把这种分布叫做Zipf's Law。企业，城市方面
已经有了一些很好的经济学模型解释得不错了。
不知道邮票上有没有点深刻的原因？
【在 t*******y 的大作中提到】
: In my opinion, about 95% of the stamps in the world has almost no value (
: below $5). About 4.5% has some value ($5-20), and the top 0.5% which has
: value of $20-$10000+, accounts for 95% value of all stamps.
: Check China/PRC stamps and this figure is generally applied.
: With that, that's usually no need to check the values of most stamps,
: especially those came in a album. I have a number of international albums
: with 3000-5000 stamps, mint and used, from 1880-1970, total worth is less
: than $50.
: A f
被下载的文件等等，数学上把这种分布叫做Zipf's Law。企业，城市方面
已经有了一些很好的经济学模型解释得不错了。
不知道邮票上有没有点深刻的原因？
【在 t*******y 的大作中提到】
: In my opinion, about 95% of the stamps in the world has almost no value (
: below $5). About 4.5% has some value ($5-20), and the top 0.5% which has
: value of $20-$10000+, accounts for 95% value of all stamps.
: Check China/PRC stamps and this figure is generally applied.
: With that, that's usually no need to check the values of most stamps,
: especially those came in a album. I have a number of international albums
: with 3000-5000 stamps, mint and used, from 1880-1970, total worth is less
: than $50.
: A f
t*y
6 楼
My view about the reason behind this is that stamp prices are determined by the stmap's popularity, quantity and the collecting community's purchasing power.
1. We can treat the entire collecting community as one person (the super collector), all the existing stamps were collected by this person, and all the stamp albums as one gigantic one. Now one new collector starts the hobby and he/she need to purchase a stamp from the super collector. Rationally the super collector will sell any stamps in
1. We can treat the entire collecting community as one person (the super collector), all the existing stamps were collected by this person, and all the stamp albums as one gigantic one. Now one new collector starts the hobby and he/she need to purchase a stamp from the super collector. Rationally the super collector will sell any stamps in
h*s
7 楼
E说得不错，后来我想了想其实不复杂，
假如所有的邮票都在1000美元上下浮动，这样只要有一小部分人对着最高价的拿几枚争抢
就会把价钱抬上去，珍邮也由此而定义了，这样distribution也就出来了。
这就好比如果大家初始都有100million左右的财富，那样较低的那部分人就依然是穷人，
最高的几个人成了富人，贫富标准也是这样出来的。
但是firm和city的size有个更有趣的现象，就是最大的firm/city的size是第二大的一
倍，第三大的两倍，。。。这个不知道在邮票里面是不是也这样。最值钱的是第二值钱的
一倍，。。。
by
collector), all the existing stamps were collected by this person, and all
the stamp albums as one gigantic one. Now one new collector starts the hobby
and he/she need to purcha
Revenue), thus create higher demand, thus their pr
【在 t*******y 的大作中提到】
: My view about the reason behind this is that stamp prices are determined by the stmap's popularity, quantity and the collecting community's purchasing power.
: 1. We can treat the entire collecting community as one person (the super collector), all the existing stamps were collected by this person, and all the stamp albums as one gigantic one. Now one new collector starts the hobby and he/she need to purchase a stamp from the super collector. Rationally the super collector will sell any stamps in
假如所有的邮票都在1000美元上下浮动，这样只要有一小部分人对着最高价的拿几枚争抢
就会把价钱抬上去，珍邮也由此而定义了，这样distribution也就出来了。
这就好比如果大家初始都有100million左右的财富，那样较低的那部分人就依然是穷人，
最高的几个人成了富人，贫富标准也是这样出来的。
但是firm和city的size有个更有趣的现象，就是最大的firm/city的size是第二大的一
倍，第三大的两倍，。。。这个不知道在邮票里面是不是也这样。最值钱的是第二值钱的
一倍，。。。
by
collector), all the existing stamps were collected by this person, and all
the stamp albums as one gigantic one. Now one new collector starts the hobby
and he/she need to purcha
Revenue), thus create higher demand, thus their pr
【在 t*******y 的大作中提到】
: My view about the reason behind this is that stamp prices are determined by the stmap's popularity, quantity and the collecting community's purchasing power.
: 1. We can treat the entire collecting community as one person (the super collector), all the existing stamps were collected by this person, and all the stamp albums as one gigantic one. Now one new collector starts the hobby and he/she need to purchase a stamp from the super collector. Rationally the super collector will sell any stamps in
t*y
8 楼
PRC stamps kind of follow the Zipf's Law. Here's the top ranking ones:
1. C94M $5500
2. C94i $2500, W10 $2000
3. C106M $1500, W7 $1400, W1 $1200, S61M $1200
4. T41M $800, C86M $600, C94 $600, T46 $550, W2/W3/S57/S4 orginal $500
5. J41M $350, S38/S44 $300, S56/S61 $250, T28M $220 ...
There're about 50 set of stamps over $100 in PRC, the total value of them is $24K. Giving the total sets of PRC stamps now in 1000, this 0.5% accounts for 60% of the total PRC stamp value. If we consider the quantity
1. C94M $5500
2. C94i $2500, W10 $2000
3. C106M $1500, W7 $1400, W1 $1200, S61M $1200
4. T41M $800, C86M $600, C94 $600, T46 $550, W2/W3/S57/S4 orginal $500
5. J41M $350, S38/S44 $300, S56/S61 $250, T28M $220 ...
There're about 50 set of stamps over $100 in PRC, the total value of them is $24K. Giving the total sets of PRC stamps now in 1000, this 0.5% accounts for 60% of the total PRC stamp value. If we consider the quantity
h*s
9 楼
very interesting, let me do some research and get back to you.
Where did you get this price list?
is $24K. Giving the total sets of PRC stamps now in 1000, this 0.5% accounts
for 60% of the total PRC stamp value. If we consider the quantity factor (
recent stamps were printe
【在 t*******y 的大作中提到】
: PRC stamps kind of follow the Zipf's Law. Here's the top ranking ones:
: 1. C94M $5500
: 2. C94i $2500, W10 $2000
: 3. C106M $1500, W7 $1400, W1 $1200, S61M $1200
: 4. T41M $800, C86M $600, C94 $600, T46 $550, W2/W3/S57/S4 orginal $500
: 5. J41M $350, S38/S44 $300, S56/S61 $250, T28M $220 ...
: There're about 50 set of stamps over $100 in PRC, the total value of them is $24K. Giving the total sets of PRC stamps now in 1000, this 0.5% accounts for 60% of the total PRC stamp value. If we consider the quantity
Where did you get this price list?
is $24K. Giving the total sets of PRC stamps now in 1000, this 0.5% accounts
for 60% of the total PRC stamp value. If we consider the quantity factor (
recent stamps were printe
【在 t*******y 的大作中提到】
: PRC stamps kind of follow the Zipf's Law. Here's the top ranking ones:
: 1. C94M $5500
: 2. C94i $2500, W10 $2000
: 3. C106M $1500, W7 $1400, W1 $1200, S61M $1200
: 4. T41M $800, C86M $600, C94 $600, T46 $550, W2/W3/S57/S4 orginal $500
: 5. J41M $350, S38/S44 $300, S56/S61 $250, T28M $220 ...
: There're about 50 set of stamps over $100 in PRC, the total value of them is $24K. Giving the total sets of PRC stamps now in 1000, this 0.5% accounts for 60% of the total PRC stamp value. If we consider the quantity
mb
10 楼
我个人认为对于任何top 1％的东西(邮票，文物书画，公司，富人财产，天才。。。）
是没有理性的模型去解释的
争抢
人，
钱的
【在 h*s 的大作中提到】
: E说得不错，后来我想了想其实不复杂，
: 假如所有的邮票都在1000美元上下浮动，这样只要有一小部分人对着最高价的拿几枚争抢
: 就会把价钱抬上去，珍邮也由此而定义了，这样distribution也就出来了。
: 这就好比如果大家初始都有100million左右的财富，那样较低的那部分人就依然是穷人，
: 最高的几个人成了富人，贫富标准也是这样出来的。
: 但是firm和city的size有个更有趣的现象，就是最大的firm/city的size是第二大的一
: 倍，第三大的两倍，。。。这个不知道在邮票里面是不是也这样。最值钱的是第二值钱的
: 一倍，。。。
:
: by
是没有理性的模型去解释的
争抢
人，
钱的
【在 h*s 的大作中提到】
: E说得不错，后来我想了想其实不复杂，
: 假如所有的邮票都在1000美元上下浮动，这样只要有一小部分人对着最高价的拿几枚争抢
: 就会把价钱抬上去，珍邮也由此而定义了，这样distribution也就出来了。
: 这就好比如果大家初始都有100million左右的财富，那样较低的那部分人就依然是穷人，
: 最高的几个人成了富人，贫富标准也是这样出来的。
: 但是firm和city的size有个更有趣的现象，就是最大的firm/city的size是第二大的一
: 倍，第三大的两倍，。。。这个不知道在邮票里面是不是也这样。最值钱的是第二值钱的
: 一倍，。。。
:
: by
h*s
11 楼
恰恰相反，有不少东西都有很漂亮的经济学模型了。
比如穷人和富人收入的distribution
【在 mb 的大作中提到】
: 我个人认为对于任何top 1％的东西(邮票，文物书画，公司，富人财产，天才。。。）
: 是没有理性的模型去解释的
:
: 争抢
: 人，
: 钱的
比如穷人和富人收入的distribution
【在 mb 的大作中提到】
: 我个人认为对于任何top 1％的东西(邮票，文物书画，公司，富人财产，天才。。。）
: 是没有理性的模型去解释的
:
: 争抢
: 人，
: 钱的
eu
12 楼
haha popularity
小时候看到美国DISNEY邮票漂亮的不得了, 买了N套, 又不贵, 读中学的时候拿了两套
到班上去, 女生都围着转啊:)
by the stmap's popularity, quantity and the collecting community's
purchasing power.
collector), all the existing stamps were collected by this person, and all
the stamp albums as one gigantic one. Now one new collector starts the hobby
and he/she need to purchase a stamp from the super collector. Rationally
the super collector will sell any stamps in proportion of its quantity. This
is the quantity factor.
Revenue), t
【在 t*******y 的大作中提到】
: My view about the reason behind this is that stamp prices are determined by the stmap's popularity, quantity and the collecting community's purchasing power.
: 1. We can treat the entire collecting community as one person (the super collector), all the existing stamps were collected by this person, and all the stamp albums as one gigantic one. Now one new collector starts the hobby and he/she need to purchase a stamp from the super collector. Rationally the super collector will sell any stamps in
小时候看到美国DISNEY邮票漂亮的不得了, 买了N套, 又不贵, 读中学的时候拿了两套
到班上去, 女生都围着转啊:)
by the stmap's popularity, quantity and the collecting community's
purchasing power.
collector), all the existing stamps were collected by this person, and all
the stamp albums as one gigantic one. Now one new collector starts the hobby
and he/she need to purchase a stamp from the super collector. Rationally
the super collector will sell any stamps in proportion of its quantity. This
is the quantity factor.
Revenue), t
【在 t*******y 的大作中提到】
: My view about the reason behind this is that stamp prices are determined by the stmap's popularity, quantity and the collecting community's purchasing power.
: 1. We can treat the entire collecting community as one person (the super collector), all the existing stamps were collected by this person, and all the stamp albums as one gigantic one. Now one new collector starts the hobby and he/she need to purchase a stamp from the super collector. Rationally the super collector will sell any stamps in
z*9
13 楼
what about T4 price?
is $24K. Giving the total sets of PRC stamps now in 1000, this 0.5% accounts
for 60% of the total PRC stamp value. If we consider the quantity factor (
recent stamps were printed in 10s of millions), then all the CSWN stamps
with high end TJs will become the top 5% and accouts for 95% of all PRC
stamp value.
【在 t*******y 的大作中提到】
: PRC stamps kind of follow the Zipf's Law. Here's the top ranking ones:
: 1. C94M $5500
: 2. C94i $2500, W10 $2000
: 3. C106M $1500, W7 $1400, W1 $1200, S61M $1200
: 4. T41M $800, C86M $600, C94 $600, T46 $550, W2/W3/S57/S4 orginal $500
: 5. J41M $350, S38/S44 $300, S56/S61 $250, T28M $220 ...
: There're about 50 set of stamps over $100 in PRC, the total value of them is $24K. Giving the total sets of PRC stamps now in 1000, this 0.5% accounts for 60% of the total PRC stamp value. If we consider the quantity
is $24K. Giving the total sets of PRC stamps now in 1000, this 0.5% accounts
for 60% of the total PRC stamp value. If we consider the quantity factor (
recent stamps were printed in 10s of millions), then all the CSWN stamps
with high end TJs will become the top 5% and accouts for 95% of all PRC
stamp value.
【在 t*******y 的大作中提到】
: PRC stamps kind of follow the Zipf's Law. Here's the top ranking ones:
: 1. C94M $5500
: 2. C94i $2500, W10 $2000
: 3. C106M $1500, W7 $1400, W1 $1200, S61M $1200
: 4. T41M $800, C86M $600, C94 $600, T46 $550, W2/W3/S57/S4 orginal $500
: 5. J41M $350, S38/S44 $300, S56/S61 $250, T28M $220 ...
: There're about 50 set of stamps over $100 in PRC, the total value of them is $24K. Giving the total sets of PRC stamps now in 1000, this 0.5% accounts for 60% of the total PRC stamp value. If we consider the quantity
d*i
14 楼
S4
accounts
【在 z***9 的大作中提到】
: what about T4 price?
:
: is $24K. Giving the total sets of PRC stamps now in 1000, this 0.5% accounts
: for 60% of the total PRC stamp value. If we consider the quantity factor (
: recent stamps were printed in 10s of millions), then all the CSWN stamps
: with high end TJs will become the top 5% and accouts for 95% of all PRC
: stamp value.
accounts
【在 z***9 的大作中提到】
: what about T4 price?
:
: is $24K. Giving the total sets of PRC stamps now in 1000, this 0.5% accounts
: for 60% of the total PRC stamp value. If we consider the quantity factor (
: recent stamps were printed in 10s of millions), then all the CSWN stamps
: with high end TJs will become the top 5% and accouts for 95% of all PRC
: stamp value.
mb
15 楼
我想你可能没看清我说什么，请看原贴
）
你回贴里的例子好像有点答非所问的感觉。就邮票而言应该没有你前面假设的
‘第一大是第二大的2倍之类的现象’
PRC发行的正票里面最贵的5个应该是
1. 梅M 4。5万
2。普5 3。5
3。梅无齿 2。3
4。文10 2。3
5。文7 1。7
其余在1万以上的票还有4－5个都不存在2倍现象
世界最富的人也不存在2倍现象
1. 62 billion
2. 60
3. 58
4. 45
5. 43
最新数据参考
http://www.forbes.com/lists/2008/03/05/richest-people-billionaires-billionaire
s08-cx_lk_0305billie_land.html
财富500公司也不存在2倍现象
1.351 billion
2.347
3.318
最新数据参考
http://money.cnn.com/galleries/2007/fortune/0707/gallery.global500_top25.fortu
ne/index.html
还有前面techsavvy贴里说的0。5％的
）
你回贴里的例子好像有点答非所问的感觉。就邮票而言应该没有你前面假设的
‘第一大是第二大的2倍之类的现象’
PRC发行的正票里面最贵的5个应该是
1. 梅M 4。5万
2。普5 3。5
3。梅无齿 2。3
4。文10 2。3
5。文7 1。7
其余在1万以上的票还有4－5个都不存在2倍现象
世界最富的人也不存在2倍现象
1. 62 billion
2. 60
3. 58
4. 45
5. 43
最新数据参考
http://www.forbes.com/lists/2008/03/05/richest-people-billionaires-billionaire
s08-cx_lk_0305billie_land.html
财富500公司也不存在2倍现象
1.351 billion
2.347
3.318
最新数据参考
http://money.cnn.com/galleries/2007/fortune/0707/gallery.global500_top25.fortu
ne/index.html
还有前面techsavvy贴里说的0。5％的
mb
16 楼
上面忘了说一句
统计结果得要囊括所有数据，比如上面如果去掉普5这套票 2倍这个结论好像适用于
PRC票，但。。。 统计要完全。不能只‘用’能被模型解释的数据
【在 mb 的大作中提到】
: 我想你可能没看清我说什么，请看原贴
: ）
: 你回贴里的例子好像有点答非所问的感觉。就邮票而言应该没有你前面假设的
: ‘第一大是第二大的2倍之类的现象’
: PRC发行的正票里面最贵的5个应该是
: 1. 梅M 4。5万
: 2。普5 3。5
: 3。梅无齿 2。3
: 4。文10 2。3
: 5。文7 1。7
统计结果得要囊括所有数据，比如上面如果去掉普5这套票 2倍这个结论好像适用于
PRC票，但。。。 统计要完全。不能只‘用’能被模型解释的数据
【在 mb 的大作中提到】
: 我想你可能没看清我说什么，请看原贴
: ）
: 你回贴里的例子好像有点答非所问的感觉。就邮票而言应该没有你前面假设的
: ‘第一大是第二大的2倍之类的现象’
: PRC发行的正票里面最贵的5个应该是
: 1. 梅M 4。5万
: 2。普5 3。5
: 3。梅无齿 2。3
: 4。文10 2。3
: 5。文7 1。7
h*s
17 楼
Zipf's law只是一个大概，第一比第二大一倍只是通俗的说法，
实际上只要找到足够的data，拟合出来在zipf的参数范围内就可以了。
【在 mb 的大作中提到】
: 我想你可能没看清我说什么，请看原贴
: ）
: 你回贴里的例子好像有点答非所问的感觉。就邮票而言应该没有你前面假设的
: ‘第一大是第二大的2倍之类的现象’
: PRC发行的正票里面最贵的5个应该是
: 1. 梅M 4。5万
: 2。普5 3。5
: 3。梅无齿 2。3
: 4。文10 2。3
: 5。文7 1。7
实际上只要找到足够的data，拟合出来在zipf的参数范围内就可以了。
【在 mb 的大作中提到】
: 我想你可能没看清我说什么，请看原贴
: ）
: 你回贴里的例子好像有点答非所问的感觉。就邮票而言应该没有你前面假设的
: ‘第一大是第二大的2倍之类的现象’
: PRC发行的正票里面最贵的5个应该是
: 1. 梅M 4。5万
: 2。普5 3。5
: 3。梅无齿 2。3
: 4。文10 2。3
: 5。文7 1。7
mb
18 楼
富人和大公司的数据和一倍这个模型相差很远，误差应该50％以上吧
【在 h*s 的大作中提到】
: Zipf's law只是一个大概，第一比第二大一倍只是通俗的说法，
: 实际上只要找到足够的data，拟合出来在zipf的参数范围内就可以了。
【在 h*s 的大作中提到】
: Zipf's law只是一个大概，第一比第二大一倍只是通俗的说法，
: 实际上只要找到足够的data，拟合出来在zipf的参数范围内就可以了。
h*s
19 楼
我一开始的帖子就没提到富人，
firm size符合这个law在很多国家都验证了，70年代就被人发现了
到了95年才出来第一个模型解释，去年出了一个更加漂亮的模型
【在 mb 的大作中提到】
: 富人和大公司的数据和一倍这个模型相差很远，误差应该50％以上吧
firm size符合这个law在很多国家都验证了，70年代就被人发现了
到了95年才出来第一个模型解释，去年出了一个更加漂亮的模型
【在 mb 的大作中提到】
: 富人和大公司的数据和一倍这个模型相差很远，误差应该50％以上吧
t*y
20 楼
hbs must be math major ;-)
There're many ways to find PRC stamp prices. Here's a few sites:
1. eBay (Worldwide)
2. pudongstamps.com (Shanghai)
3. zhaoonline.com auction results (Shanghai)
4. e1988.com (Beijing)
5. Michael Roger's price list (US)
6. fidelitytc.com (US)
6. major HongKong and Taiwan auction houses, such as philachina.com,
bennettstamps.com, spink.com, jbull.com, www.panlung.com.tw, etc
accounts
【在 h*s 的大作中提到】
: very interesting, let me do some research and get back to you.
: Where did you get this price list?
:
: is $24K. Giving the total sets of PRC stamps now in 1000, this 0.5% accounts
: for 60% of the total PRC stamp value. If we consider the quantity factor (
: recent stamps were printe
There're many ways to find PRC stamp prices. Here's a few sites:
1. eBay (Worldwide)
2. pudongstamps.com (Shanghai)
3. zhaoonline.com auction results (Shanghai)
4. e1988.com (Beijing)
5. Michael Roger's price list (US)
6. fidelitytc.com (US)
6. major HongKong and Taiwan auction houses, such as philachina.com,
bennettstamps.com, spink.com, jbull.com, www.panlung.com.tw, etc
accounts
【在 h*s 的大作中提到】
: very interesting, let me do some research and get back to you.
: Where did you get this price list?
:
: is $24K. Giving the total sets of PRC stamps now in 1000, this 0.5% accounts
: for 60% of the total PRC stamp value. If we consider the quantity factor (
: recent stamps were printe
mb
21 楼
请拿出相应数据支持你的结论
【在 h*s 的大作中提到】
: 我一开始的帖子就没提到富人，
: firm size符合这个law在很多国家都验证了，70年代就被人发现了
: 到了95年才出来第一个模型解释，去年出了一个更加漂亮的模型
【在 h*s 的大作中提到】
: 我一开始的帖子就没提到富人，
: firm size符合这个law在很多国家都验证了，70年代就被人发现了
: 到了95年才出来第一个模型解释，去年出了一个更加漂亮的模型
h*s
22 楼
Gabaix (1995) Quarterly Journal of Economics
从他那里可以找到很多文献和数据。
此人大牛，此文是他的job market paper，四年毕业去了MIT
去年被挖到NYU。9个月base salary 最保守估计25万$外加funding无数
身为经济学家，经常发science和nature。
今年这个新模型是我一个同学，就不cite了。
没兴趣看论文的话，wiki “Zipf's Law”
【在 mb 的大作中提到】
: 请拿出相应数据支持你的结论
从他那里可以找到很多文献和数据。
此人大牛，此文是他的job market paper，四年毕业去了MIT
去年被挖到NYU。9个月base salary 最保守估计25万$外加funding无数
身为经济学家，经常发science和nature。
今年这个新模型是我一个同学，就不cite了。
没兴趣看论文的话，wiki “Zipf's Law”
【在 mb 的大作中提到】
: 请拿出相应数据支持你的结论
t*y
23 楼
We are here for fun, no need to be so serious and debate the theories.
I think hbs's idea is very interesting, and the market roughly follows it. Consider the complexity, worldwide reach and the dynamic nature of the stamp market, I have to say the zipf's law has some truth. If a theory can explain 80% of the fact, it is already very good ;-) Taking about using a single formula to explain millions of people's social and economical behavior ;-)
MB sorry I forgot about R5, probably because I never
I think hbs's idea is very interesting, and the market roughly follows it. Consider the complexity, worldwide reach and the dynamic nature of the stamp market, I have to say the zipf's law has some truth. If a theory can explain 80% of the fact, it is already very good ;-) Taking about using a single formula to explain millions of people's social and economical behavior ;-)
MB sorry I forgot about R5, probably because I never
h*s
24 楼
I asked a math prof who has expertise in Zipf's Law and he told me it is
hard to come up with a explanation. But he was unware of Gabaix and my
fellow's papers. We may consider this phenomena again.
I am busy these days so I won't delve into it shortly. But I will let
you guys know if I have something.
2004 Nobel winner in economics explained the driving forces of business
cycle by tech productivity which account for 70% of cycle. Prior to them,
everyone believed that business cycle is caused by
【在 t*******y 的大作中提到】
: We are here for fun, no need to be so serious and debate the theories.
: I think hbs's idea is very interesting, and the market roughly follows it. Consider the complexity, worldwide reach and the dynamic nature of the stamp market, I have to say the zipf's law has some truth. If a theory can explain 80% of the fact, it is already very good ;-) Taking about using a single formula to explain millions of people's social and economical behavior ;-)
: MB sorry I forgot about R5, probably because I never
hard to come up with a explanation. But he was unware of Gabaix and my
fellow's papers. We may consider this phenomena again.
I am busy these days so I won't delve into it shortly. But I will let
you guys know if I have something.
2004 Nobel winner in economics explained the driving forces of business
cycle by tech productivity which account for 70% of cycle. Prior to them,
everyone believed that business cycle is caused by
【在 t*******y 的大作中提到】
: We are here for fun, no need to be so serious and debate the theories.
: I think hbs's idea is very interesting, and the market roughly follows it. Consider the complexity, worldwide reach and the dynamic nature of the stamp market, I have to say the zipf's law has some truth. If a theory can explain 80% of the fact, it is already very good ;-) Taking about using a single formula to explain millions of people's social and economical behavior ;-)
: MB sorry I forgot about R5, probably because I never
t*y
25 楼
Richest people who have $20+ billion. Interestingly there's a big gap between 50 and 30. And on average first tier $50billion is about double the average of second tier $25billion.
However, from 20 billion and under this gap disappears. A more continuous series is shown.
Rank Name Citizenship Age Net Worth ($bil) Residence
1 William Gates III United States 51 56.0
United States
2 Warren Buffett United States 76 52.0 United States
3 C
However, from 20 billion and under this gap disappears. A more continuous series is shown.
Rank Name Citizenship Age Net Worth ($bil) Residence
1 William Gates III United States 51 56.0
United States
2 Warren Buffett United States 76 52.0 United States
3 C
h*s
26 楼
hehe,
I wish someday your name will be listed.
BTW, if next time someone offer you something that you do not want,
refer him to me, my new job pays very well.
between 50 and 30. And on average first tier $50billion is about double the
average of second tier $25billion.
series is shown.
【在 t*******y 的大作中提到】
: Richest people who have $20+ billion. Interestingly there's a big gap between 50 and 30. And on average first tier $50billion is about double the average of second tier $25billion.
: However, from 20 billion and under this gap disappears. A more continuous series is shown.
: Rank Name Citizenship Age Net Worth ($bil) Residence
: 1 William Gates III United States 51 56.0
: United States
: 2 Warren Buffett United States 76 52.0 United States
: 3 C
I wish someday your name will be listed.
BTW, if next time someone offer you something that you do not want,
refer him to me, my new job pays very well.
between 50 and 30. And on average first tier $50billion is about double the
average of second tier $25billion.
series is shown.
【在 t*******y 的大作中提到】
: Richest people who have $20+ billion. Interestingly there's a big gap between 50 and 30. And on average first tier $50billion is about double the average of second tier $25billion.
: However, from 20 billion and under this gap disappears. A more continuous series is shown.
: Rank Name Citizenship Age Net Worth ($bil) Residence
: 1 William Gates III United States 51 56.0
: United States
: 2 Warren Buffett United States 76 52.0 United States
: 3 C
mb
27 楼
概率统计我也学过zeta分布也早知道
我想问题的焦点在于到底哪些数据符合zeta分布
说句不中听的话，如果‘选择性’的选取数据的话，任何数据中可以找出符合任何
模型的数据来。
关于firm size，有数据可查，下面列了employees of America's Largest Private
Companies
http://www.forbes.com/lists/2007/21/biz_privates07_Americas-Largest-Private-Companies_Employees.html
你可以看看是否符合zeta分布。
如果你有其他数据支持你的结论你可以列出，远比空洞地陈述有力的多。
【在 h*s 的大作中提到】
: Gabaix (1995) Quarterly Journal of Economics
: 从他那里可以找到很多文献和数据。
: 此人大牛，此文是他的job market paper，四年毕业去了MIT
: 去年被挖到NYU。9个月base salary 最保守估计25万$外加funding无数
: 身为经济学家，经常发science和nature。
: 今年这个新模型是我一个同学，就不cite了。
: 没兴趣看论文的话，wiki “Zipf's Law”
我想问题的焦点在于到底哪些数据符合zeta分布
说句不中听的话，如果‘选择性’的选取数据的话，任何数据中可以找出符合任何
模型的数据来。
关于firm size，有数据可查，下面列了employees of America's Largest Private
Companies
http://www.forbes.com/lists/2007/21/biz_privates07_Americas-Largest-Private-Companies_Employees.html
你可以看看是否符合zeta分布。
如果你有其他数据支持你的结论你可以列出，远比空洞地陈述有力的多。
【在 h*s 的大作中提到】
: Gabaix (1995) Quarterly Journal of Economics
: 从他那里可以找到很多文献和数据。
: 此人大牛，此文是他的job market paper，四年毕业去了MIT
: 去年被挖到NYU。9个月base salary 最保守估计25万$外加funding无数
: 身为经济学家，经常发science和nature。
: 今年这个新模型是我一个同学，就不cite了。
: 没兴趣看论文的话，wiki “Zipf's Law”
mb
28 楼
hehe，就像我前面贴里说的那样，我不认为理论能解释top 1%的数据一样
哪怕80%的准确性都没有
特别是对于近期中国邮票价格，更是没有任何规律可言
10年前普5是梅张价格的近2倍，现在呢？这次涨价前两者价格又是什么关系呢？
还是那句话 ‘选择性'地找出来的数据适合任何模型。就像NASA连续几十年监听
太空中的信号一样，总会找到一些’有规律‘的声音。但说明什么呢？
Consider the complexity, worldwide reach and the dynamic nature of the
stamp market, I have to say the zipf's law has some truth. If a theory can
explain 80% of the fact, it i
(I was once offered a complete PRC collection, but I asked the owner to
take R5 out to save some money ;-). It's value is now $4500, close
【在 t*******y 的大作中提到】
: We are here for fun, no need to be so serious and debate the theories.
: I think hbs's idea is very interesting, and the market roughly follows it. Consider the complexity, worldwide reach and the dynamic nature of the stamp market, I have to say the zipf's law has some truth. If a theory can explain 80% of the fact, it is already very good ;-) Taking about using a single formula to explain millions of people's social and economical behavior ;-)
: MB sorry I forgot about R5, probably because I never
哪怕80%的准确性都没有
特别是对于近期中国邮票价格，更是没有任何规律可言
10年前普5是梅张价格的近2倍，现在呢？这次涨价前两者价格又是什么关系呢？
还是那句话 ‘选择性'地找出来的数据适合任何模型。就像NASA连续几十年监听
太空中的信号一样，总会找到一些’有规律‘的声音。但说明什么呢？
Consider the complexity, worldwide reach and the dynamic nature of the
stamp market, I have to say the zipf's law has some truth. If a theory can
explain 80% of the fact, it i
(I was once offered a complete PRC collection, but I asked the owner to
take R5 out to save some money ;-). It's value is now $4500, close
【在 t*******y 的大作中提到】
: We are here for fun, no need to be so serious and debate the theories.
: I think hbs's idea is very interesting, and the market roughly follows it. Consider the complexity, worldwide reach and the dynamic nature of the stamp market, I have to say the zipf's law has some truth. If a theory can explain 80% of the fact, it is already very good ;-) Taking about using a single formula to explain millions of people's social and economical behavior ;-)
: MB sorry I forgot about R5, probably because I never
h*s
29 楼
I do not know why you are so mad at this.
What I have said is that there are some widely accepted data belong to
some category are distributed according to Zipf's Law and there are
rational forces behind it. We never said this is also applicable to
stamp prices. What we said is just a simple CONJECTURE! We could be
wrong.
Of course you can even doubt about the former like firm or city sizes,
that's your freedom.
Take a look at the followings:
Gabaix, Xavier (August 1999). Zipf's Law for Cities
【在 mb 的大作中提到】
: 概率统计我也学过zeta分布也早知道
: 我想问题的焦点在于到底哪些数据符合zeta分布
: 说句不中听的话，如果‘选择性’的选取数据的话，任何数据中可以找出符合任何
: 模型的数据来。
: 关于firm size，有数据可查，下面列了employees of America's Largest Private
: Companies
: http://www.forbes.com/lists/2007/21/biz_privates07_Americas-Largest-Private-Companies_Employees.html
: 你可以看看是否符合zeta分布。
: 如果你有其他数据支持你的结论你可以列出，远比空洞地陈述有力的多。
What I have said is that there are some widely accepted data belong to
some category are distributed according to Zipf's Law and there are
rational forces behind it. We never said this is also applicable to
stamp prices. What we said is just a simple CONJECTURE! We could be
wrong.
Of course you can even doubt about the former like firm or city sizes,
that's your freedom.
Take a look at the followings:
Gabaix, Xavier (August 1999). Zipf's Law for Cities
【在 mb 的大作中提到】
: 概率统计我也学过zeta分布也早知道
: 我想问题的焦点在于到底哪些数据符合zeta分布
: 说句不中听的话，如果‘选择性’的选取数据的话，任何数据中可以找出符合任何
: 模型的数据来。
: 关于firm size，有数据可查，下面列了employees of America's Largest Private
: Companies
: http://www.forbes.com/lists/2007/21/biz_privates07_Americas-Largest-Private-Companies_Employees.html
: 你可以看看是否符合zeta分布。
: 如果你有其他数据支持你的结论你可以列出，远比空洞地陈述有力的多。
h*s
30 楼
You misunderstood our points from the beginning, nobody is focus on top 1%
data, the most interesting thing is to see if everyone is born equal, then
what can drive some dimension of them to be distributed unevenly in the
long run, so that we have top 1%, top 20%, ...., bottom 20%, bottom 1%.
I just said, in my profession, it is well recoginzed that these can be
explained rationally and we do have some very successful models explaining
the distribution of INCOME (this one is not distributed in Z
【在 mb 的大作中提到】
: hehe，就像我前面贴里说的那样，我不认为理论能解释top 1%的数据一样
: 哪怕80%的准确性都没有
: 特别是对于近期中国邮票价格，更是没有任何规律可言
: 10年前普5是梅张价格的近2倍，现在呢？这次涨价前两者价格又是什么关系呢？
: 还是那句话 ‘选择性'地找出来的数据适合任何模型。就像NASA连续几十年监听
: 太空中的信号一样，总会找到一些’有规律‘的声音。但说明什么呢？
:
: Consider the complexity, worldwide reach and the dynamic nature of the
: stamp market, I have to say the zipf's law has some truth. If a theory can
: explain 80% of the fact, it i
data, the most interesting thing is to see if everyone is born equal, then
what can drive some dimension of them to be distributed unevenly in the
long run, so that we have top 1%, top 20%, ...., bottom 20%, bottom 1%.
I just said, in my profession, it is well recoginzed that these can be
explained rationally and we do have some very successful models explaining
the distribution of INCOME (this one is not distributed in Z
【在 mb 的大作中提到】
: hehe，就像我前面贴里说的那样，我不认为理论能解释top 1%的数据一样
: 哪怕80%的准确性都没有
: 特别是对于近期中国邮票价格，更是没有任何规律可言
: 10年前普5是梅张价格的近2倍，现在呢？这次涨价前两者价格又是什么关系呢？
: 还是那句话 ‘选择性'地找出来的数据适合任何模型。就像NASA连续几十年监听
: 太空中的信号一样，总会找到一些’有规律‘的声音。但说明什么呢？
:
: Consider the complexity, worldwide reach and the dynamic nature of the
: stamp market, I have to say the zipf's law has some truth. If a theory can
: explain 80% of the fact, it i
mb
31 楼
呵呵,首先没有人mad。
仔细看看之前的讨论，是否和你帖子里的话矛盾
We never said this is also applicable to stamp prices.
What we said is just a simple CONJECTURE! We could be wrong.
如果你意识到'We could be wrong.'那么就不要再用zeta分布套stamp price或是
firm size或是穷人和富人收入的distribution。如果在某个特定的假设下的特定
时间里的某个特定的数据符合zeta分布，将这些特定条件列出才是科学的做法。
就像NASA说的几十年来我们分析了几十亿条数终于据找到一条好像有规律的信号。。
虽然这句话啥也说明不了，至少人家的态度是科学的。
下面是之前的讨论：
hbs said:
说得挺好，这个现象在很多领域都存在，比如城市的大小，单词使用频率，
被下载的文件等等，数学上把这种分布叫做Zipf's Law。企业，城市方面
已经有了一些很好的经济学模型解释得不错了。
不知道邮票上有没有点深刻的原因？
hbs said:
但
【在 h*s 的大作中提到】
: I do not know why you are so mad at this.
: What I have said is that there are some widely accepted data belong to
: some category are distributed according to Zipf's Law and there are
: rational forces behind it. We never said this is also applicable to
: stamp prices. What we said is just a simple CONJECTURE! We could be
: wrong.
: Of course you can even doubt about the former like firm or city sizes,
: that's your freedom.
: Take a look at the followings:
: Gabaix, Xavier (August 1999). Zipf's Law for Cities
仔细看看之前的讨论，是否和你帖子里的话矛盾
We never said this is also applicable to stamp prices.
What we said is just a simple CONJECTURE! We could be wrong.
如果你意识到'We could be wrong.'那么就不要再用zeta分布套stamp price或是
firm size或是穷人和富人收入的distribution。如果在某个特定的假设下的特定
时间里的某个特定的数据符合zeta分布，将这些特定条件列出才是科学的做法。
就像NASA说的几十年来我们分析了几十亿条数终于据找到一条好像有规律的信号。。
虽然这句话啥也说明不了，至少人家的态度是科学的。
下面是之前的讨论：
hbs said:
说得挺好，这个现象在很多领域都存在，比如城市的大小，单词使用频率，
被下载的文件等等，数学上把这种分布叫做Zipf's Law。企业，城市方面
已经有了一些很好的经济学模型解释得不错了。
不知道邮票上有没有点深刻的原因？
hbs said:
但
【在 h*s 的大作中提到】
: I do not know why you are so mad at this.
: What I have said is that there are some widely accepted data belong to
: some category are distributed according to Zipf's Law and there are
: rational forces behind it. We never said this is also applicable to
: stamp prices. What we said is just a simple CONJECTURE! We could be
: wrong.
: Of course you can even doubt about the former like firm or city sizes,
: that's your freedom.
: Take a look at the followings:
: Gabaix, Xavier (August 1999). Zipf's Law for Cities
mb
32 楼
我想我没理解错你说的'恰恰相反‘的含义吧？
发信人: hbs (Tokyo's Rain), 信区: Collectibles
标 题: Re: 南美邮票
发信站: BBS 未名空间站 (Thu Mar 13 19:38:27 2008), 转信
恰恰相反，有不少东西都有很漂亮的经济学模型了。
比如穷人和富人收入的distribution
【在 mb 的大作中提到】
: 我个人认为对于任何top 1％的东西(邮票，文物书画，公司，富人财产，天才。。。）
: 是没有理性的模型去解释的
:
: 争抢
: 人，
: 钱的
发信人: hbs (Tokyo's Rain), 信区: Collectibles
标 题: Re: 南美邮票
发信站: BBS 未名空间站 (Thu Mar 13 19:38:27 2008), 转信
恰恰相反，有不少东西都有很漂亮的经济学模型了。
比如穷人和富人收入的distribution
【在 mb 的大作中提到】
: 我个人认为对于任何top 1％的东西(邮票，文物书画，公司，富人财产，天才。。。）
: 是没有理性的模型去解释的
:
: 争抢
: 人，
: 钱的
h*s
33 楼
你挑一句话说明我肯定了stamp prices符合Zipf's Law出来吧，
真是佩服你的阅读理解能力。
firm size的问题学术界做了几十年了，都是几十年的横截面数据，
你一个07年forbes的数据就全部否定了？
人家都不是科学做法，只有你和NASA的态度是科学的？
【在 mb 的大作中提到】
: 呵呵,首先没有人mad。
: 仔细看看之前的讨论，是否和你帖子里的话矛盾
: We never said this is also applicable to stamp prices.
: What we said is just a simple CONJECTURE! We could be wrong.
: 如果你意识到'We could be wrong.'那么就不要再用zeta分布套stamp price或是
: firm size或是穷人和富人收入的distribution。如果在某个特定的假设下的特定
: 时间里的某个特定的数据符合zeta分布，将这些特定条件列出才是科学的做法。
: 就像NASA说的几十年来我们分析了几十亿条数终于据找到一条好像有规律的信号。。
: 虽然这句话啥也说明不了，至少人家的态度是科学的。
: 下面是之前的讨论：
真是佩服你的阅读理解能力。
firm size的问题学术界做了几十年了，都是几十年的横截面数据，
你一个07年forbes的数据就全部否定了？
人家都不是科学做法，只有你和NASA的态度是科学的？
【在 mb 的大作中提到】
: 呵呵,首先没有人mad。
: 仔细看看之前的讨论，是否和你帖子里的话矛盾
: We never said this is also applicable to stamp prices.
: What we said is just a simple CONJECTURE! We could be wrong.
: 如果你意识到'We could be wrong.'那么就不要再用zeta分布套stamp price或是
: firm size或是穷人和富人收入的distribution。如果在某个特定的假设下的特定
: 时间里的某个特定的数据符合zeta分布，将这些特定条件列出才是科学的做法。
: 就像NASA说的几十年来我们分析了几十亿条数终于据找到一条好像有规律的信号。。
: 虽然这句话啥也说明不了，至少人家的态度是科学的。
: 下面是之前的讨论：
t*y
34 楼
hey don't forget to quote my earlier comment: We are here for fun, no need
to be so serious and debate the theories ;-)
to be so serious and debate the theories ;-)
mb
35 楼
兄弟，你好像火气嫌大了点。
之前你的帖子里陈述观点都是用‘we’，我想‘we’是指你自己和techsavvy。
那说明你赞同techsavvy的观点。techsavvy是认为票价符合zipf的。
现在你撇开techsavvy开始用‘我’来说事。如果这么钻字眼的话，‘你’的确没有
明说票价符合zipf.但所有说的话都直接想要支持techsavvy的观点，不是吗？
不符zipf规律的firm size何止几十上百，forbes的数据只是给你找个信得过的source
举个反例。想要更多反例的话我可以再给你找几十个。
最后，到底是我的理解力差还是。。。呵呵，明者自明。这个话题就到此为止吧
【在 h*s 的大作中提到】
: 你挑一句话说明我肯定了stamp prices符合Zipf's Law出来吧，
: 真是佩服你的阅读理解能力。
: firm size的问题学术界做了几十年了，都是几十年的横截面数据，
: 你一个07年forbes的数据就全部否定了？
: 人家都不是科学做法，只有你和NASA的态度是科学的？
之前你的帖子里陈述观点都是用‘we’，我想‘we’是指你自己和techsavvy。
那说明你赞同techsavvy的观点。techsavvy是认为票价符合zipf的。
现在你撇开techsavvy开始用‘我’来说事。如果这么钻字眼的话，‘你’的确没有
明说票价符合zipf.但所有说的话都直接想要支持techsavvy的观点，不是吗？
不符zipf规律的firm size何止几十上百，forbes的数据只是给你找个信得过的source
举个反例。想要更多反例的话我可以再给你找几十个。
最后，到底是我的理解力差还是。。。呵呵，明者自明。这个话题就到此为止吧
【在 h*s 的大作中提到】
: 你挑一句话说明我肯定了stamp prices符合Zipf's Law出来吧，
: 真是佩服你的阅读理解能力。
: firm size的问题学术界做了几十年了，都是几十年的横截面数据，
: 你一个07年forbes的数据就全部否定了？
: 人家都不是科学做法，只有你和NASA的态度是科学的？
h*s
36 楼
这里大家都是明眼人。
不是你火气大，techsavvy也不会回帖说we are here for fun not for debate。
techsavvy的观点我只是说了very interesting。你理解成了我赞同。
我的确有兴趣深入想一想，但以我受过的经济学训练，我必须先去看看其他国家
的邮票价格，假如中国邮票的的确确符合Zipf，而其他国家的不符合，那也没必要
做下去了，这个问题也没意义了。
我最不明白的是学术界普遍接受的firm size符合Zipf这个事情（这个问题最早
是和一个经济学Nobel得主联系在一起的，所以很多人follow)，你那么大意见。
我还是那句话，firm size的事情学术界做了几十年，你如果能找几十个反例出来，
然后把你的观点发到science（差一点的journal也行）上再来表态不迟（我等五年
吧。science很喜欢这个话题，有data的话可以很快发表，经济学期刊即使
是很好的paper也得等个三四年）。
source
【在 mb 的大作中提到】
: 兄弟，你好像火气嫌大了点。
: 之前你的帖子里陈述观点都是用‘we’，我想‘we’是指你自己和techsavvy。
: 那说明你赞同techsavvy的观点。techsavvy是认为票价符合zipf的。
: 现在你撇开techsavvy开始用‘我’来说事。如果这么钻字眼的话，‘你’的确没有
: 明说票价符合zipf.但所有说的话都直接想要支持techsavvy的观点，不是吗？
: 不符zipf规律的firm size何止几十上百，forbes的数据只是给你找个信得过的source
: 举个反例。想要更多反例的话我可以再给你找几十个。
: 最后，到底是我的理解力差还是。。。呵呵，明者自明。这个话题就到此为止吧
不是你火气大，techsavvy也不会回帖说we are here for fun not for debate。
techsavvy的观点我只是说了very interesting。你理解成了我赞同。
我的确有兴趣深入想一想，但以我受过的经济学训练，我必须先去看看其他国家
的邮票价格，假如中国邮票的的确确符合Zipf，而其他国家的不符合，那也没必要
做下去了，这个问题也没意义了。
我最不明白的是学术界普遍接受的firm size符合Zipf这个事情（这个问题最早
是和一个经济学Nobel得主联系在一起的，所以很多人follow)，你那么大意见。
我还是那句话，firm size的事情学术界做了几十年，你如果能找几十个反例出来，
然后把你的观点发到science（差一点的journal也行）上再来表态不迟（我等五年
吧。science很喜欢这个话题，有data的话可以很快发表，经济学期刊即使
是很好的paper也得等个三四年）。
source
【在 mb 的大作中提到】
: 兄弟，你好像火气嫌大了点。
: 之前你的帖子里陈述观点都是用‘we’，我想‘we’是指你自己和techsavvy。
: 那说明你赞同techsavvy的观点。techsavvy是认为票价符合zipf的。
: 现在你撇开techsavvy开始用‘我’来说事。如果这么钻字眼的话，‘你’的确没有
: 明说票价符合zipf.但所有说的话都直接想要支持techsavvy的观点，不是吗？
: 不符zipf规律的firm size何止几十上百，forbes的数据只是给你找个信得过的source
: 举个反例。想要更多反例的话我可以再给你找几十个。
: 最后，到底是我的理解力差还是。。。呵呵，明者自明。这个话题就到此为止吧
相关阅读
kongling 由 otrmajia 授予 Collectibles 俱乐部权力meilefu 由 MB 授予 Collectibles 俱乐部权力三只小狮子ebay老卖家的假清票?大家都去哪里淘宝阿？一件单色釉瓷器，始终没有搞懂这两幅画也许可以给Happyy同学一些参考贴个刺刀裸金币铂币怎么储存？各位大牛，帮看一下这枚银币值不值为什么有人说这个是假的关于铂金proteind 由 otrmajia 授予 Collectibles 俱乐部权力再贴两个冷门一点的八国联军德国军邮wodashu 由 otrmajia 授予 Collectibles 俱乐部权力Mars 由 otrmajia 授予 Collectibles 俱乐部权力一张清末民初的地契刚收到的中奥加厚panda deal of the day!再出些熊猫金币