Oil Market Takes Stock After Wild Trading in Election Aftermath
The oil market is taking a breather.
A day after Donald Trump’s shock U.S. presidential election victory Wednesday whipsawed prices on some of the heaviest trading volumes on record, U.S. futures hovered above $45 a barrel amid a rally in commodities from metals to grains. Crude pared earlier gains after the International Energy Agency said Thursday that prices may retreat amid “relentless global supply growth” unless the Organization of Petroleum Exporting Countries enacts significant output cuts.
Please register on futures.io to view futures trading content such as post attachment(s), image(s), and screenshot(s).
Traders are weighing the implications of the Republican presiding over a country that consumes more oil than any other and is one of the biggest producers too. Trump has promised independence from OPEC and some of his energy policies include opening federal lands for energy production and freeing up offshore areas to development. While investors took comfort from a conciliatory acceptance speech, a surge in U.S. crude stockpiles served as a reminder of the massive oversupply looming over the market.
“There is still a lot of uncertainty,” Toby Lawson, head of global markets for Australia at Societe Generale SA, said in a Bloomberg television interview. “We’ve got a bit of time to go through, to get some certainty around what the actual policies are but the direction indicates quite clearly that a Trump administration will be looking to fiscal policy to drive growth.”
Interesting question. I don't know, but consider this ... If we say the market has a 50% chance of being up or down, then the chance of 9 down years out of 10 is (0.5)^9*(0.5)^1. But there's 10 ways we can get 9 out of 10, so the actual probability is 10*(0.5)^9*(0.5)^1 which is approximately 0.0098% which equates to 1 in 102 days. So approximately every 102 days, this should happen by chance alone.
Excuse my lack of statistical knowledge, but am I correct that your formula telling us the chances of being down 9 out of 10 days?
It's down over a 4 day period so does your formula apply to a 4 day period?
Since 2000 RBz down 14 out of 16.
I can find seasonal patterns looking at seasonal charts but it would help if I knew the why on the seasonal movement so that I can predict if that might happen again this year. Some of them I have no idea.
For example grains are usually down in Sep. That is probably because US harvest is happening and risk is gone for weather affecting crop once it is harvested.
The following 2 users say Thank You to ron99 for this post:
Yes correct, my bad. It's a probability of 0.0098 which is 0.98%.
It could be 9 out 10 days, or 9 out of 10 years.
I was just saying 'down over a 4 day period is 50/50, and then 9 years out of 10'.
(16*15)*(0.5^16) = 0.0037 or 1 in 273
Note that while there are only 10 ways we can arrange 9 downs, and 1 up out of 10, there are 16*15 ways we can arrange 14 downs and 2 ups out of 16. Wikipedia-Permutation's
There are good reasons for seasonal patterns to occur, be they fundamental or behavioral. I wasn't saying that's not the case here, I was just pointing out that at least twice a year you should be able to find a similar situation, purely due to chance. Obviously if there is a reason behind the pattern, theres a good chance it occurs again. If it's just chance though, it's only 50/50 it happens again.
The following user says Thank You to SMCJB for this post:
No. It's saying that once every 102 trading DAYS we should randomly find a day where the market has been up 9 out of the last 10 years on that day. So given 250 trading days a year, we would expect to see these 9 out 10 pattern 2.45 times per year purely based upon chance.
How about this.
I'm going to use R to generate a random table of data (called m), with 10 rows, representing 10 years, and 250 columns, representing 250 trading days. If the random number is greater than 0.5 we will round it up to 1 (ie an up day), and less than 0.5 down to 0 (ie a down day). If we then sum each column it will show us how many days were up vs down for that day in the last 10 years. I'll then create a table to show the frequency of each outcome.
*runif is the R function to generate a random number from the uniform distribution
So in this example we got 1 out of 250 occurrences where the market was up 9 years out 10, but 4 out of 250 occurrences where it was down 9 years out of 10.
Running this a second time will yield different results because it will use a different random numbers used. To illustrate this I will run it in a loop 10 times, each time using a different seed for the random number generator. (Note that since in this case I'm setting the seed if somebody else runs the same code they should get identical results as they will get the same random numbers as I did.)
Now you can see our 9 out of 10 pattern occurred 0, 5, 6, 1, 4, 1, 0 , 2, 1, 3 times respectively which has a mean of 2.3 vs our expectation of 2.44
The following 2 users say Thank You to SMCJB for this post: