|
|
|||
| |
 |
Ed
Seykota's FAQ
|
| Home | Send Mail to FAQ | Resources | Ground Rules | FAQ Index |
TT_Chartbook
|
| FAQ
December 01-10, 2016 <-- back | down | forward --> |
| Contributors Say | Ed Says | |||
| Dec 10, 2016 Trump Ed, I wonder what you think of Trump winning the election? |
Thank you for raising this issue. Now that we have campaign rhetoric behind us, you might consider monitoring what policies the president and the congress actually implement. Protectionism: generally lowers the standard of living for consumers and may lead to war. Lower Taxes and Reduce Government: generally raises the standard of living and motivates business expansion. One policy that might have a profound effect on the economy does not seem to appear on anyone's agenda: return to the gold standard. |
|||
| Dec 9,
2016 Occasional Trading System Dear Ed, Thank you for keeping on updating your website with FAQ. I am thinking on the results of one trading system and I would be happy to know your thoughts on this. One system is a very long trend following systems. Same rules to generate long and short trades and to get out of these trades. The trading system is tested in one market over the last 30 years and it produces the following results: Profit Factor: 9.64 Ratio Average Win/loss: 9.64 Percent Profitable: 50% Average Monthly Profit: 0.31% of the contract value Max Drawdown (Trade to Trade, NOT intraday): 5.43% of the contract value Total number of trades: 16 It is true that these results have been obtained over a very long period of time, but would you consider statistically significant a result obtained with 16 trades only? On the other hand, if we used a short term trading system that has generated more than 500 trades, even if we had a definitely lower profit factor could we consider the results statistically more significant? I really thank you for any thought on the subject. |
Thank
you for raising this issue. Yes, 16 trades do not give you much statistical verification - if you don't have anything else. Perhaps you have something else. You might consider getting more data by testing your long-term 16-trade system in other markets. From the data you have, if the short-term system differs from the long-term system only in the time parameters - and if the short-term system also shows profits, then you might have more confidence in the long-term system. You might consider that statistical relevance presuppose a large sample size - and that even a statistically profitable system indicates very little about your chances on any one particular trade.
|
|||
| Dec 8,
2016 Luck and Trading Ed, On your list of Important Investors, I notice that "Every single person on this list is incredibly smart. But they were also fortunate to be born when they were" Every once in a while an unlucky guy gets lucky. See: https://en.wikipedia.org/wiki/John_D._Arnold |
Thank
you for raising this issue. I think that success has a lot to do with the right place at the right time. A lot has to do with recognizing opportunity and jumping into it. A lot has to do with character and attitude. I think success and happiness do not correlate linearly - and that the latter has more to do with love and grace.
|
|||
| Dec 5,
2016 Important Investors Timeline Ed, I see you on a timeline of important investors.
|
Thank
you for sending me the link. |
|||
| Dec 4,
2016 Performance Versus Stop Tightness Dear Ed, First of all thanks for taking the time to help others in a lot of subjects, not only trading related. Thanks also, for taking the time to read this email. I've been researching several trend following systems, trying to have the least possible parameters to avoid over fitting. I'm trying to have only a long-term period resistance which determine entry signals, which during optimization, seems to be very stable and another one, which is the stop loss distance. The stop loss parameter is influenced by a variable which is multiplied by ATR to determine stop loss level (adjvar - explained below). The larger this variable is, the larger the stop loss distance. I've attached a picture which shows Compounded Return/Max. Drawdown in Z axis, Stop "tightness" in "adjvar" axis and risk per trade in the other axis.
From optimization (see attached picture) for very tight stops the CAR/MDD is high, but then we see a drop in performance, and for larger stop distance we see an improvement again. (doesn't seem a robust parameter) What I've also found is that in this system for larger stops, we get higher CAR and higher Drawdowns, for the same CAR/MDD Ratio. Do you have some insights from your experience, on the performance of systems in relation with stop loss tightness, that can help in researching the effects of stop distance in systems performance? Inline image 2 Thanks again for your time Kind Regards, |
Thank
you for raising
this issue. I do not see much in the way of a "sweet spot" on your graphic. You might consider running a series of tests to generate surfaces with different values of "skid" or transaction cost, ranging from 0% to 100% of the range on the day of trade. You might find your "tighter" systems turn out much more sensitive to transaction costs than your "looser" ones. |
|||
| Dec 1,
2016 Robots at Work Ed, In this video of an auto plant production line, I finally see some people at about the 7-minute mark. |
Thank
you for sharing
the video. |
|||
| Dec 1,
2016 Popular Thinking: Protectionism Ed, Contrary to popular thought, the US is not growing slowly because technology is stealing jobs. It is not growing slowly because productivity isn't improving. It is not stuck in a secular stagnation because of China, Mexico or inequality. The US is growing slowly because it is carrying the burden of a massive government. All government spending is financed by borrowing or taxing from the private sector. The more the government spends and the more it redistributes the more it "crowds out" economic growth. The bigger the government, the slower the economy, jobs and incomes grow. Federal Government spending, excluding defense, has risen from around 6% of GDP to over 17% of GDP in the past 60 years. If the Trump Administration … really wants to improve living conditions for the average American, not just in Rust Belt states, but everywhere, what it needs to do is reduce the size and scope of the federal government. |
Thank
you for sharing
your analysis. Popular thought does not always reflect the lessons of history. For example, during the final stages of an inflation-money-printing blow off, the public screams for more and more money printing so they can keep up with inflation.
|
|||
| Dec 1,
2016 Hyper Inflation Ed, The Venezuelan Bolivar has just experienced its most devastating and destructive short-term decline in history! On October 24th, the cost of USD$1 in Venezuela was Bs. 1,221.87, but the announcement of a 40% minimum wage hike has caused the cost of USD$1 in Venezuela to explode in recent days to Bs. 4,538.46. Over the past month, the Venezuelan Bolivar has lost a stunning -64.64% of its purchasing power against the U.S. dollar, its largest month-to-month decline in history. For comparison, its median month-to-month decline over the past 5 years has been only -5.95%. This gives Venezuela an implied annualized price inflation rate of over 26 million percent! |
Thank
you for posting
this information. For more on the dynamics on hyperinflation - and what you can and cannot do about it - see my book, Govopoly in the 39th Day, above. |
|||
| Dec 1,
2016 So Much Information Hi Ed, Greetings. I have started going through your website and found it interesting and useful. I am based out of India and have a deep interest in trading and believe in personal excellence in all aspects of life. I have a question to ask you. I am currently trying to understand markets and I see lot of black box systems, statistical algorithms, advances in machine learning and all. How does one get a perspective on learning with so much information around. |
Thank
you for raising
this issue. I wonder if you can tell me what you mean by "getting a perspective on learning." |
|||
| To Top of Page |
