7月第24天
昨日阅读2小时,累计阅读371小时
1.今天你阅读到的有价值的全文内容链接
推荐: 《亿万:围剿华尔街大白鲨》 作者:Sheelah Kolhatkar
史蒂文·科恩改变了华尔街。他和对冲基金业的同道先锋,既不铺铁路,也不建工厂,更不发明新技术。相反,他们只是通过投机,利用胜多败少的市场押注,就狂赚数十亿美元--他们不仅因此取得了惊人的个人财富,还在社会上获得了可怕的影响力。 https://bbs.pinggu.org/thread-6530181-1-1.html
2.今天你阅读到的有价值的内容段落摘录
今天继续读:《Definitive Guide to Position Sizing》:Chapter 2
Risk(R) and R-Multiples (Answers for 10 questions of R-multiples,and more.)
……
Answers: Be sure you understand these answers before moving ahead in this workbook. In each case, I indicate 1R. I then divide the profit or loss by 1R to determine the R-multiple. It’s that simple.
- 1R is 2 USD. Your loss per share is 9 USD, so you have a 4.5R loss.
- 1R is 4 USD. Your profit per share is 40 USD, so you have a 10R profit.
- 1R is 4 USD. Your profit per share is 5 USD, so you have a 1.25R profit.
- 1R is 5 USD. Your loss per share is 60 USD, so you have a 12R loss. You never want to let this happen.
- 1R is 1 USD. Your profit per share is 20 USD, so you have a 20R profit. You want this to happen all the time.
- 1R is 12.5 USD. Your loss per share is 2 USD, so you have a 0.16R loss. This is the sort of loss you want. Some people might argue that you’ve allowed a profit to turn into a loss. However, the key is you have followed your rules.
- 1R is 1.50 USD or half the valus of the option. Your profit is 9 USD, which is a 6R profit.
- 1R is 1.50 USD. Your total loss is 4.50 USD, so you have a 3R loss.
- 1R is 10 cents. Your total profit is 1.50 USD, so you have a 15R gain. Notice that it was simple to determine your R-multiple from a single bushel of wheat because you just needed the ratio.
- 1R is 35 cents. Your profit is 11.13 USD (i.e., 51.48 USD less your cost of 40.35 USD = 11.13 USD ). If you divide 11.13 USD by 35 cents, you get a profit of 31.8R. Again, this is the kind of profit you want. Notice that your profit was only about 26% of the initial cost of the stock, but your result is a huge R-multiple because your 1R value was so small.
Using Your Total Risk to Keep Track of Your R-Multiples
It can get quite complex to keep track of the risk per unit and the profit or loss per unit. In addition, there are also transaction costs involved that won’t get figured into your profit or loss per share. As a result, an easier way to determine the R-multiple distribution of your trades is to use the total initial risk and the total profit or loss (after costs) to determine your R-multiples.
Let’s say that you have a 100,000 USD account and you want to keep your total risk per position to about 1% of your account value or 1,000 USD. Here’s what a sample of trades might look like.
- You buy a stock at 40 USD and plan to exit if it drops to 38 USD. You buy 500 shares, which at a risk of 2USD per share gives you a total risk of 1,000 USD. However, the stock goes to 37 USD and then gaps down five points at the open the next day. You get out as soom as you can at 31 USD. Your total loss is 9 USD per share times 500 shares or 4,500 USD. You also had a transaction cost of 24 USD, making your total loss 4,524USD.
Your initial total risk is 1,000USD. Your total loss was 4,500USD, so you had a 4.5R loss. Wasn’t this the same answer you got for the first example above? It should have been. However, your total loss, including transaction costs, was 4,524USD. Thus, your actual R-multiple loss was 4.524R. Notice how the only difference between using total risk and risk per share is that with total risk you can include all of your costs in the R-multiple, which makes it a little more accurate.
By the way, you could also use the total cost to determine 1R. However, this step makes your calculations even more complex and your results would be slightly different. As a result, I recommend that you simply subtract your total costs from your profit or add it to your losses, as I did in the above example. You’ll find that it is much simpler and you’ve still included the total cost in your R-multiples.
……
3.今天你阅读到的有价值信息的自我思考点评感想
更为详细的算法是增加了交易手续费的计算,当然为了更快掌握,初期只是建议增加在损失计算的时候。实际盈利中还应当增加税的计算,需要根据实际交易对于A股的交易计算一下详细的R乘数。 如果交易的量比较大的话,交易手续费和税对于成本核算还是非常重要的,尤其是在频率较高的交易系统中(当然,频率会和R直接相关)。
这也需要一个模型,只是对于R乘数的理解更进了一点,模型应该开始构思….
确实是需要研究透彻,算账才会更清晰。