g*****g
发帖数: 34805 | 1 我个人认为200日线的方向更为重要。
现在更像02年底到03年初的熊市反弹，
幅度很大，速度很快，但跌起来的时候
更快。但3月份都不一定是底，乐观地估计
也要重回8000，形成头肩底。当然，市场
完全可以先跑到9000。 |
m**********r
发帖数: 887 | 2 A Quantitative Approach to Tactical Asset Allocation
文件太大，我把pdf链接贴这里了
http://trendfollowing.com/whitepaper/CMT-Simple.pdf |
m**********r
发帖数: 887 | 3 The system is as follows:
BUY RULE
Buy when monthly price > 10-month SMA.
SELL RULE
Sell and move to cash when monthly price < 10-month SMA.
1. All entry and exit prices are on the day of the signal at the close.
2. All data series are total return series including dividends, updated
monthly.
3. Cash returns are estimated with 90-day commercial paper, and margin rates
(for leveraged models to be
discussed later) are estimated with the broker call rate.
4. Taxes, commissions, and slippage are exc |
c*******a
发帖数: 653 | 4 俺是这么盼望的啊～～
【在 g*****g 的大作中提到】
: 我个人认为200日线的方向更为重要。
: 现在更像02年底到03年初的熊市反弹，
: 幅度很大，速度很快，但跌起来的时候
: 更快。但3月份都不一定是底，乐观地估计
: 也要重回8000，形成头肩底。当然，市场
: 完全可以先跑到9000。
|
o*****c
发帖数: 241 | 5 ran this strategy on spx since 1927.
got sharpe ratio of 0.399. the best parameter is crossing 15 months SMA with
sharpe of 0.459, which is still not high enough to be used in real trading.
rates
【在 m**********r 的大作中提到】
: The system is as follows:
: BUY RULE
: Buy when monthly price > 10-month SMA.
: SELL RULE
: Sell and move to cash when monthly price < 10-month SMA.
: 1. All entry and exit prices are on the day of the signal at the close.
: 2. All data series are total return series including dividends, updated
: monthly.
: 3. Cash returns are estimated with 90-day commercial paper, and margin rates
: (for leveraged models to be
|
h*******y
发帖数: 864 | 6 赞专业.
with
【在 o*****c 的大作中提到】
: ran this strategy on spx since 1927.
: got sharpe ratio of 0.399. the best parameter is crossing 15 months SMA with
: sharpe of 0.459, which is still not high enough to be used in real trading.
:
: rates
|
b*****e
发帖数: 1125 | 7 Take trading cost you may find the sharp close to 0.
with
【在 o*****c 的大作中提到】
: ran this strategy on spx since 1927.
: got sharpe ratio of 0.399. the best parameter is crossing 15 months SMA with
: sharpe of 0.459, which is still not high enough to be used in real trading.
:
: rates
|
o*****c
发帖数: 241 | 8 it is based on monthly SPX data. you can almost ignore the trading cost if
trading ES/SPY going forward.
【在 b*****e 的大作中提到】
: Take trading cost you may find the sharp close to 0.
:
: with
|
b*****e
发帖数: 1125 | 9 Try 20bps trading cost for each buy and sell, let see what the result. It
should be a fairly quick work in excel. |
u**o
发帖数: 4652 | |
|
|
h*******y
发帖数: 864 | 11 Some form of risk-adjusted return calculation. The issue with sharpe ratio
is that it uses volatility as a proxy of the inherent risk.
The sharpe ratio of S&P 500 is 0.4 so if the goal is to beat S&P 500 it
needs to be significantly higher than that of S&P 500.
【在 u**o 的大作中提到】
: what is sharpe ratio ?
|
h*******y
发帖数: 864 | 12 By the way, I didn't mean that the investment is good as long as its sharpe
ratio is significantly higher than S&P 500. LTCM before its bust has a
sharpe ratio well over 4. So to each of his own...
【在 h*******y 的大作中提到】
: Some form of risk-adjusted return calculation. The issue with sharpe ratio
: is that it uses volatility as a proxy of the inherent risk.
: The sharpe ratio of S&P 500 is 0.4 so if the goal is to beat S&P 500 it
: needs to be significantly higher than that of S&P 500.
|
s********n
发帖数: 1962 | 13 all TA has this problem: they work until they don't.
sharpe
【在 h*******y 的大作中提到】
: By the way, I didn't mean that the investment is good as long as its sharpe
: ratio is significantly higher than S&P 500. LTCM before its bust has a
: sharpe ratio well over 4. So to each of his own...
|
b*****e
发帖数: 1125 | 14 Should that be a problem for any strategy not just TA? |
m**********r
发帖数: 887 | |
o*****c
发帖数: 241 | 16 if you are really in a trading desk, you should know slippage+commission
cost should be less than 3bps for trading long-term index futures.
ok. let's use the nonsense 20 bps.
sharpe is 0.392 for crossing 10 sma and 0.456 for crossing 15 sma.
【在 b*****e 的大作中提到】
: Try 20bps trading cost for each buy and sell, let see what the result. It
: should be a fairly quick work in excel.
|
o*****c
发帖数: 241 | 17 it should be less than 0.15 if you buy and hold sp since 1927. i didn't
consider dividend though.
【在 h*******y 的大作中提到】
: Some form of risk-adjusted return calculation. The issue with sharpe ratio
: is that it uses volatility as a proxy of the inherent risk.
: The sharpe ratio of S&P 500 is 0.4 so if the goal is to beat S&P 500 it
: needs to be significantly higher than that of S&P 500.
|
b*****e
发帖数: 1125 | 18 Since 1927 up to when?
My naive testing shows 10 SMA cumulative P&L is beaten by buy and hold from
Oct 1928 to May 2009 (pre sample reserved for moving average). Of course i
ignored rebate, fincing and trading cost etc.
with
【在 o*****c 的大作中提到】
: ran this strategy on spx since 1927.
: got sharpe ratio of 0.399. the best parameter is crossing 15 months SMA with
: sharpe of 0.459, which is still not high enough to be used in real trading.
:
: rates
|
o*****c
发帖数: 241 | 19 did we talk about sharpe ratio? do you know the difference between sharpe
and PnL?
so you like the profit chart of buy&hold below???
from
【在 b*****e 的大作中提到】
: Since 1927 up to when?
: My naive testing shows 10 SMA cumulative P&L is beaten by buy and hold from
: Oct 1928 to May 2009 (pre sample reserved for moving average). Of course i
: ignored rebate, fincing and trading cost etc.
:
: with
|
b*****e
发帖数: 1125 | 20 I don't have any preference on either strategy. I just said what I observed.
Btw, I think your sharp ratio calculation maybe wrong, especially after
taking trading cost.
【在 o*****c 的大作中提到】
: did we talk about sharpe ratio? do you know the difference between sharpe
: and PnL?
: so you like the profit chart of buy&hold below???
:
: from
|
|
|
o*****c
发帖数: 241 | 21 then get the right sharpe for us, pls.
if avg profit is over 9% per trade, why do you care about the 20bps or
several bps trading cost? that doesn't make any sense.that is not a high
frequency strategy, right?
observed.
【在 b*****e 的大作中提到】
: I don't have any preference on either strategy. I just said what I observed.
: Btw, I think your sharp ratio calculation maybe wrong, especially after
: taking trading cost.
|
g*****g
发帖数: 34805 | 22 I don't think trade cost is the problem.
This kind of long term indicators show up once a year at most.
But I am not sure if you've counted the return when the money
sits aside. Unlike B&H, you are not fully in at all time.
【在 o*****c 的大作中提到】
: then get the right sharpe for us, pls.
: if avg profit is over 9% per trade, why do you care about the 20bps or
: several bps trading cost? that doesn't make any sense.that is not a high
: frequency strategy, right?
:
: observed.
|
b*****e
发帖数: 1125 | 23 I will calculate the vol for months in position only, exclude cash holding
months.
【在 o*****c 的大作中提到】
: then get the right sharpe for us, pls.
: if avg profit is over 9% per trade, why do you care about the 20bps or
: several bps trading cost? that doesn't make any sense.that is not a high
: frequency strategy, right?
:
: observed.
|
