***Official*** 2017 Stock Market Thread

[Imp]

Everything is awesome.

 

[FelixDeCat]

And Gold goes up despite higher rates. Go figure.

 

[FelixDeCat]

In 1999, I remember everyone was astonished how far and how fast the market rose because of the internet hysteria. By December 1999 everyone was leery but the market kept rising. I did not sell. It started to fluctuate in February 2000 and by April 2000 we started the free fall that did not end until 2002, when you could have gotten any stock for a song.

Apple was $19 after they announced gigantic losses and stock crashed 40% in one day. Almost six years later is was $300 after the start of ipod / iphone then was $89 a year later after the global credit crises. So many chances....

 

[MomentsofSanity]

Thought I would throw this out for people since most investors have no idea it's going on but it's driving us nuts on the order entry side.

https://www.sec.gov/oiea/investor-alerts-bulletins/ia_ticksize.html

SEC is running a pilot on some low volume securities. Essentially they are requiring that orders for the securities they select for the pilot be entered in $0.05 intervals. You can still fill out at odd numbers but the orders have to be in multiples of 5 cents. More info and list of securities impacted etc... at link.

 

[bookem dano]

I think people are listening to Buffett when he says if you don't know what you are doing, just got some index funds....
http://www.nasdaq.com/article/dow-jones-index-etf-most-popular-so-far-this-year-cm747191

The fund, based on the venerable and—to some–controversial Dow Jones Industrial Average Index, is the most popular ETF so far this year, and has taken in $2.56 billion in new assets since the beginning of the year. That’s the most of any U.S.-listed ETF in 2017...... with $4.5 billion in new assets since Nov. 8, and is up more than 12%, outpacing QQQ (up 10%) and SPY (up 9.7%) during the same period.​

 

[Charmonium]

That's a little worrying. Retail investors are the first to shit their pants when things go sideways. Then you have redemptions and funds having to sell to cover them. And not all of the assets they hold are all that liquid so you can get some dramatic drops if funds have to do significant liquidations. Index funds probably aren't affected by this but smaller, more specialized funds probably would be.

I guess we'll see what happens next month with the Fed. If retail investors really are a big part of this market rally I would expect the news of any rate increase to spark a sell off. For the pros, that info is already baked in and stocks would probably rally on the news but for the average investor who thinks he understand the markets, they might take that as a signal to sell.

 

[Imp]

I think people are listening to Buffett when he says if you don't know what you are doing, just got some index funds....

I get the feeling that even more people are listening to their in-laws and buying property...

 

[FelixDeCat]

What goes up
must come down
Spinnin' wheel
got to go round...

 

[Imp]

Every time I think about stocks, it's actually this song that comes to mind...

https://youtu.be/xKCek6_dB0M

Edit: Reading about Canadian real-estate is hilarious.

 

[FelixDeCat]

Every time I think about stocks, it's actually this song that comes to mind...

https://youtu.be/xKCek6_dB0M

Edit: Reading about Canadian real-estate is hilarious.

Well, my only reply to this "ride a painted pony....let the spinnin' wheel spin".

 

[Charmonium]

I'm dating myself but it'll probably be more like this - https://www.youtube.com/watch?v=B8gy6IIlW04 Rivers of Blood, Years of Darkness

 

[repoman0]

I started a new python project the other night ... stock market timing algorithms. I've been kicking around the idea in my head for a few years but found another site where a couple engineers came up with some models that work well on past market data and charge a ton of money for access to them. I figure, I TOO am an engineer who works with complicated data all day ... should be able to come up with a testing framework, a few models, optimize model parameters using past data, backtesting and hopefully hypothetically avoiding past market drops and recessions ... then applying to the current market over time and hope the models are robust enough that they work on out of testing set data points.

Learning python is a nice bonus since I can't get free matlab anymore except on my work laptop and I hate my work laptop.

 

[KB]

I started a new python project the other night ... stock market timing algorithms. I've been kicking around the idea in my head for a few years but found another site where a couple engineers came up with some models that work well on past market data and charge a ton of money for access to them. I figure, I TOO am an engineer who works with complicated data all day ... should be able to come up with a testing framework, a few models, optimize model parameters using past data, backtesting and hopefully hypothetically avoiding past market drops and recessions ... then applying to the current market over time and hope the models are robust enough that they work on out of testing set data points.

Learning python is a nice bonus since I can't get free matlab anymore except on my work laptop and I hate my work laptop.

Good luck. I have thought of doing this myself, but so much of it seems to be greed and fear, things that are hard to measure.
Remember when the market just keeps going up its easy to test your algo and think its doing it right. And back-testing data only tests the past, not the future.

One thing I always ask myself, if the algos of these engineers were so great why would they sell access and not just keep them to themselves making tons of money for themselves.

 

[repoman0]

Good luck. I have thought of doing this myself, but so much of it seems to be greed and fear, things that are hard to measure.
Remember when the market just keeps going up its easy to test your algo and think its doing it right. And back-testing data only tests the past, not the future.

Agreed with you there. The hope is that there are examples in the past for what greed and fear look like in terms of market and economic data and the models effectively capture it. If I can come up with something that signals a move to bonds or cash in all past recessions and large market drops, I would feel fairly confident in its future ability to do so. We shall see though, at the least it's good coding and data analysis practice.

One thing I always ask myself, if the algos of these engineers were so great why would they sell access and not just keep them to themselves making tons of money for themselves.

The one in particular I'm thinking of is imarketsignals.com. It looks like they do it both for enjoyment and making money -- they have a quite a few different models and analyses and have chosen to publish some of them and keep some of them proprietary. I don't find it too unusual to try to get both active and passive income out of their work. The active portion is risk free income while the investment portion could still fail, no matter how good -- and it's on a longer time scale.

Here's their market timer that got me thinking a little harder about it beyond throwing it around in my head:

http://imarketsignals.com/2015/spy-ief-market-timer/

 

[FelixDeCat]

OK, tell me why I shouldn't buy shitloads of FNMA?

The feds are done with it, right?--so that means de-reg, HIGH DONALD INTEREST RATES, poor people and middle class lose their homes--again, and so shares skyrocket, right?

right?

It looks like FNMA is going to stay government owned for sometime. Its down 25% today!

 

[Imp]

I started a new python project the other night ... stock market timing algorithms.

If you're interested, sounds like a fun thing to try... but remember that there are many investment firms out there with multiple PhD holders poached from the best universities who have done the same thing -- sounds like algo-trading. And their servers are parked beside the stock exchange's servers.

And the S&P 500 is up again! It must be DOW 40,000. Someone somewhere showed a chart of bull-markets reaching 800% and 900% runs in the 1930s and 1970s or something, we're only at like 400% now so past performance must be indicative of future returns!

 

[repoman0]

If you're interested, sounds like a fun thing to try... but remember that there are many investment firms out there with multiple PhD holders poached from the best universities who have done the same thing -- sounds like algo-trading. And their servers are parked beside the stock exchange's servers.

This idea is different from that -- no speed required, just a weekly readjustment between overall stock market, overall bond market and cash. Inputs would be last N days of market close prices, VIX values, and who knows what else. Nothing crazy, I just want something that (hopefully) semi-intelligently tells me how to allocate investments for the week, or month, or whatever in an attempt to avoid sustained drops.

 

[SketchMaster]

There is an interesting book, Automate This!, that talks about how the stock market came into the digital age. They started by splicing a stock ticker into a computer to get live feeds and got shut down multiple times because it was viewed as cheating. Now you have people paying billions of dollars trying to figure out how to run the straightest cable possible to NY; just to get that trade in 0.00000001/sec faster than the other guy.

 

[jpiniero]

I guess we'll see what happens next month with the Fed. If retail investors really are a big part of this market rally I would expect the news of any rate increase to spark a sell off. For the pros, that info is already baked in and stocks would probably rally on the news but for the average investor who thinks he understand the markets, they might take that as a signal to sell.

How high do you think the rates would have to get for companies to stop mortgaging the farm to buyback stock?

 

[FelixDeCat]

How high do you think the rates would have to get for companies to stop mortgaging the farm to buyback stock?

That will be looked back upon as one very foolish thing to have done.

 

[zinfamous]

It looks like FNMA is going to stay government owned for sometime. Its down 25% today!

 

[louis redfoot]

just picked up some whole foods shares. it's at a historic low, a sound business and quality play. in its space it is top dog. i just look at the clientele: a lot of athletes, gay guys, milfs in yoga pants. this ain't vons!

 

[zinfamous]

just picked up some whole foods shares. it's at a historic low, a sound business and quality play. in its space it is top dog. i just look at the clientele: a lot of athletes, gay guys, milfs in yoga pants. this ain't vons!

It's the Apple of groceries: pretty good quality, but 500% markup on crap you don't need but think you do.

 

[louis redfoot]

It's the Apple of groceries: pretty good quality, but 500% markup on crap you don't need but think you do.

a lot of stuff is overpriced, just gotta know how to shop. and then there are those who don't care

 

[zinfamous]

a lot of stuff is overpriced, just gotta know how to shop. and then there are those who don't care

Right; but I hope you don't think that I was suggesting that Apple was a bad buy back in the day when they stuck a 3.5mm headphone jack on a hard drive, charged $400, and called it a revolution.