Autoregressive Order one process introduction and example

This video provides an introduction to Autoregressive Order One processes, and provides an example of a process which could be modelled in this way. Check out ben-lambert.co... for course materials, and information regarding updates on each of the courses. Quite excitingly (for me at least), I am about to publish a whole series of new videos on Bayesian statistics on youtube. See here for information: ben-lambert.co... Accompanying this series, there will be a book: www.amazon.co....

[@tazzron]

It's come to the point where I've stopped looking at lecture slides and exclusively come here to learn.

[@boubougeorge1777]

That is so right 😂

[@umiddey8714]

It's 2020 and I do the same.

[@thhbere]

2021, same

[@donghoshin3460]

2022, same

[@Flush333]

2023, same

[@yaweli2968]

It’s a crime you are teaching for free and my professor is getting paid when he can’t teach. Thanks for what you do. Much appreciated.

[@samkim6933]

You are a genius~!! made me intuitively understand the material when no other books or lecturers couldn't...

[@lizrael]

Thank you so much! You have helped me understand more than any of my profs ever had.

[@_kim7271]

유가쇼크 후 가격이 원상복귀하는 과정이라니... 설명 미쳤다... 이게 ㄹㅇ 계량경제학이지...

[@Eizengoldt]

Hi cutie

[@attesaarinen4311]

Ben you da MVP

[@wanjadouglas3058]

Very nice example Lambert

[@Garet43]

This is an excellent explanation of the AR1 effect! Thank you!!

[@changlinlei7600]

Thank you sir for the clear explanation

[@Alex-uo1ef]

Great video! Helped me a lot for understanding the AR Model in the context of signal processing !

[@Highlyk]

Great vids Ben. Incredibly useful.

[@amengioio]

Thanks for making this informative video.
Should the formula of the oil price example just be, "Oilp[t] = .5*Oilp[t-1] + eps[t]" instead of "delta(Oilp[t]) = .5*delta(Oilp[t-1]) + eps[t]"?
Because if I expand the formula, it seems to be a AR(2) process.

[@tusharbharati5151]

+Ryan Zhang
I agree.
+Ben Lambert: Please take a look

[@tommarty234]

+Ryan Zhang keep in mind that delta(Oilp[t]) is the random variable being considered, not Oilp[t] - so the AR(1) model's recursive order to Oilp[t], or any other variable, is not relevant in this context. Also I think he has defined delta(Oilp[t]) as the delta from some constant, as opposed to delta between the last two values, otherwise his plot would oscillate. I realise this question is from over a year ago, Im just replying in case anyone else has the same question.

[@pedromrfernandes]

i guess with the formula in deltas the example is correct. however, with the formula in levels, the price would have to climb by an extra $5 in t+1, reaching +15$ from the period before the shock, then reaching a peak right after the change vanish. can anyone confirm?

[@ecerulm]

This video seems to be part of series, I would be nice to include a link to the series/playlist (if there is one) in the description or better yet as overlay annotation in the video itself.

[@namukayasandra5633]

Thank you.I have been having trouble understanding y applied econometrics module

[@ΓιωργοςΓαλάρης-ψ3β]

Thanks a lot for these videos! Very helpfull!!

[@mehmethikmet3439]

Ben you are a fucking legend mate.

[@iamdumbmatt]

very intuitive explanation sir

[@urabi3tube]

I have a question why do use AR or MA or ARMA model in other words why we abandon multiple regression, to be more clear I want to know why we decide to take the variable's lag as an explanatory variable?????and when also can I decide which model to build by looking to variables? please can u help me we this issue? thanks :)

[@patrickgold3616]

why is the oil price suddenly called x and not y? In previous videos, the dependent variable was always y. X here makes it seem like we are discussing an independent variable?

[@Kavafy]

It's both

[@aristalyakusuma1]

Thank you for the video. It really helps me!

[@zarulkhaliff5107]

Ben, can I use cross-section data for forecasting using AR(1)? Thanks

[@anishsharma3071]

Hello Ben,
These videos are proving themselves very helpful. Thanks for the same.
I had few questions
1. i have read this definition somewhere - "a moving average process is a linear regression of the current values of a time series against both the current and previous unobserved white noise error terms, which are random shocks". what do we mean byy saying an error term as unobserved?.
2. how we can exactly quantify unobserved white noise error term in our MA/ AR model.(have we regressed the time series and then have compared actual and modeled) ?
Please clarify, i will be thankful.

[@andrewseeran8811]

Is this a continuous "version" of a markov process?

[@ignaciosacristanlopez-brav684]

Wxcellent video like always

[@gulzameenbaloch9339]

Thank you so much

[@shairozsohail1059]

Great job!

[@dimasmukhlas3952]

Thanks man! very much help!

[@KorawichKavee]

What do you mean by iid?

[@mech_builder7998]

en.wikipedia.org/wiki/Independent_and_identically_distributed_random_variables

[@yinyuebu]

Love this!

[@SpartacanUsuals]

Hi, many thanks for your message! Best of luck with your studies. Thanks, Ben

[@ciceklale1825]

thank you for this video

[@quincyreinevergara8822]

:o

[@iremcifci5596]

your voice is disturbing for me i cannot focus

[@RoylanMartinez]

there are more videos out there, no one is asking you to watch this video

[@prateekratna]

this sucks

[@silentstorm718]

Be respectful