How has being independent served you? Have you taken steps to get there? Sending love and light wherever you are in the journey. ❤️
I got pretty unlucky with job offers when the pandemic hit. I also missed the boat and did not know how important an internship was when I was a Junior in college. I applied to over 150 companies with a 2.9 GPA and only got an offer at a construction company as a steel material and labor estimator. Am I wasting my time? I do not use any chemical engineering methods at all. Do employers care what branch you are in?
Edit: Thanks so much for the silver kind stranger!
Edit 2: And the others! You've made my day! Glad I dropped my biscuit in my tea and decided I needed answers
Hi!
A question a time series here. I have tested for unknown structural breaks in a time series, using Bai and Perron approach. Good news is that I do get the date I was expecting as a break (has to do with a policy and possibly expectations). My next point is to measure the effects the break as an aftermath. Of course, I dont have a counterfactual, but I could potentially measure the effect and calculate its impact (in $). Any ideas which model is advisable, or a paper that would be a good guide?
Hello everyone, I am testing some stocks (8) for a uni project to see if the short squeeze of the 27th of January (the GME one) caused structural breaks for some similar stocks. I have 166 observations (stock returns) which I am regressing with Nasdaq returns. the break happens at obs.113.
I have done a Chow test which revealed that 5 stocks of 8 do in fact have a structural break on date 113 (this is the observation which corresponds to the shortsqueeze, 27th of Jan).
However, when I run a Bai and Perron (1998) test of H0= 0 breaks vs H1: 1 break at unknown time, only 3 stocks seem to have structural breaks (on observation 111, 112 and 112).
Why could this be? I thought the principle of the tests was more or less the same (in my understanding B&P is more precise for "medium run").
When I ran a B&P on a known break date (113), NONE of the stocks seemed to have a structural breaks (all had p.values of above 0.60).
I also thought I would run the hypothesis of 1 vs 2 breaks, or 0 vs 2 breaks. would this be a good idea?
Thank you for your time and advice.
Are structural breaks related to a regression? what I want to verify is, if certain stocks experienced a structural break in 2021. As variables, do I need market return (the return on an index for example) and the stock return? for example, Dow Jones and AMC returns, AMC being the dependent variable?
Hi all,
I am an engineer in the UK working for a structural practise specialising in architect led design of residential and commercial projects. I have a question which I would love to get your perspective on as working architects.
The working relationship between architects and structural engineer is an important one, at least it is for the SE! Repeated work, recognition via awards, etc. often come directly through nurturing professional networks with Architects and currying favour (without overspending!).
I am interested to know what behaviours or practises you have experienced from an SE that make you want to work with them again on future work or otherwise avoid if you can help it?!
Any perspectives you can share greatly appreciated - I'm sure the point of view of Architects from outside the UK will be just as valid!
Thanks!
P.S. happy to offer the corresponding alternative perspective to anyone who asks!
What is the optimal strategy when you identify a structural break occuring very late in the series? I am using U.S. PCE data and unsurprisingly there is a significant break occuring at the outset of the pandemic. Now according to Pesaran you're supposed to ignore the time interval where the break occurs, taking only a 5-10% pre break and everything after. But there is limited data since it was so recent and my assumption is also that even with 5% it would be very biased towards the pre-break data. My own inclination is to interpolate some data during the break and go from there but I could not find any litterature to justify it and I didn't want to take too many freedoms.
So how should one handle this issue?
Thank you
In the graph representing log gdp of a country i am linking in the following imgur adress: https://imgur.com/a/Uj95ONZ, you can easily notice that there is a structural break in the variance starting from around the year 2001 that lasts up until the last year of the time series, I also detected it using statistical test.
Now I want to control for this structural break while estimating an ARDL model, to do so I created a dummy variable that is =1 when the year is superior to 2001 and 0 otherwise, which i find to be highly statistically significant when estimating the model. Is this the appropriate way to do it ? or should it just take 1 when it is the year the structural break starts?
Also, knowing that the dummy is positive for almost half of the timeseries, can this generate any bias on the rest of the parameters that I should be aware of ?
Thanks !
Surprisingly simple methods like 2 week/3 week moving average combined with business intel is working well as a short gap solution. Would love to get different views.
So I don't want to imply that the Beirut disaster didn't result in any casualties, because that's obviously not true. But I was also thinking about the asteroid that hit Chelyabinsk which resulted in a lot of broken windows but not a great deal of injuries.
Coconut oil is a great substance with many uses, but science is not in our favor this time. Thanks to chemistry and solubility and such, any latex condom is going to be weakened by contact with an oil, which may lead to holes or breaks. Non latex condoms may be fine, depending on what they're made with. You're going to the trouble of using condoms, presumably you want to be safe. Stay safe by not putting yourself at greater risk of breakage. Use condoms and water based lubricant to avoid this risk.
Happy sexing!
Edited for grammar
What I want to verify is, if certain stocks experienced a structural break in 2021. As variables, do I need market return (the return on an index for example) and the stock return? Do I need to run a regression on them first, for example, Dow Jones and AMC returns, AMC being the dependent variable? (I guess I don't entirely understand what structural breaks are)
I am doing a project in which I am meant to verify if certain stock returns experienced a structural break. Are structural breaks related to a regression? what I want to verify is, if certain stocks experienced a structural break in 2021. As variables, do I need market return (the return on an index for example) and the stock return? for example, Dow Jones and AMC returns, AMC being the dependent variable?
In the graph representing log gdp of a country i am linking in the following imgur adress: https://imgur.com/a/Uj95ONZ, you can easily notice that there is a structural break in the variance starting from around the year 2001 that lasts up until the last year of the time series, I also detected it using statistical test.
Now I want to control for this structural break while estimating an ARDL model, to do so I created a dummy variable that is =1 when the year is superior to 2001 and 0 otherwise, which i find to be highly statistically significant when estimating the model. Is this the appropriate way to do it ? or should it just take 1 when it is the year the structural break starts?
Also, knowing that the dummy is positive for almost half of the timeseries, can this generate any bias on the rest of the parameters that I should be aware of ?
Thanks !
Surprisingly simple methods like 2 week moving average combined with business intel is working well. Would love to get your views.
