#MathAI is an improved methodology of regression analysis that can be back to its main function: serving to find the accurate relationship.
The contribution of MathAI is to take each samples one-by-one adding into the regression. Alike the stepwise way by adding the independent variables one by one, an additional sample can be added into the regression. The R2 records are compared by AI to check that the additional sample belongs to the regularity of the previous samples.
Math and AI merge to advance science’s frontiers in new center. #MathAI
Hashtags: #chatGPT #MathAI #ScienceAdvancement Entities: - New Center - Math - AI - Frontiers of Science Summery: Cornell University is launching the Scientific Artificial Intelligence Center (SciAI Center) with an $11.3 million grant from the Office of Naval Research. Led by Christopher J. Earls, professor of civil and environmental engineering, the center aims to explore the collaboration between…
There was a relevant drop of Bank Credit, 17,312.855 bn dollars, updated the week of Mar. 29, 2023. The recent trend decreased on average 28 bn dollars per month.
There has been a slight drop in commercial and industrial #loans to 2.7 tn Dollars (data of Mar. 29). The latest trend direction is not changed yet, but AI notices the recent three weeks have been forming a new downward trend.
Trends of US nonfarm employees
From Jan. 2013 to Mar. 2023, there are only three best trends. So I run the nonlinear trends and get 14 trends.
The recent nonlinear trends are shown below (right).
Nonlinear trend of percentage for #layoff / #unemployment was changed to be flatter as adding the new value of Mar. 2023 that is a start of the new trend. This is not a good news. BY #mathgpt with #AI (lines combined method). 3/
The trend of percentage for #layoff / #unemployment was changed to be flatter as adding the new value of Mar. 2023 that is a start of the new trend. This is not a good news. BY #mathgpt with #AI (lines combined method). 2/
Recent trend of #layoffs has decreaseed on the average of 4.286 thousands per month since Oct. 2022. The new adding 833 thousand persons let the power of trend weaken, so the new value of Mar. 2023 will be a start of the new trend. BY #mathgpt with #AI (lines combined method). 1/
C&I loans showed the the loan value of Mar. 22, 2023 was a key time and might be a start of new trend since it let R2 of the latest trend (Feb. 8~Mar. 22) decrease.
to 03/15: R2=0.43,
to 03/22: R2=0.09.
@economics @board @laborecon @academicchatter
Compare the latest trends when the value of Mar. 2023 was added. The outcomes are calculated by MathGPT for numerical modelling (lines combined method) software. Here we can find that the US nonfarm private payrolls started stopping the decline tendency. But the downward trend is not turn around yet. 3/
Since we add the new observation, value of Mar. 2023, the change occurred on the slope of 423.749 (till Feb. 2023) to 396.399 (till Mar. 2023). 2.2/
@economics @board @laborecon
Update trend of ADP nonfram payrolls. The latest trend shows the average decrease of 387.49 (till Feb. 2023) to 323.89 (till Mar. 2023) thousands of persons per month. The nonfarm payrolls trend has declined slowly. 1.1/
The result of data which is till Feb. 2023: https://mastodon.social/@aphoebusli/110146907896918884
@board @economics @academicchatter
Recent latest trend of the delinquency rates has increased since 2021Q4, and the trend slope (0.162) is close to the one of line 5 (2016Q4~2018Q4), 0.169.
#AI財經 #AI經濟 #金融 #財經 #經濟 #finance #creditcard #loans #technology #data #dataanalytics #DataScience #mathgpt #mathai #AI
Software: MathGPT for numerical modelling (lines combined method)
US #Borrowings for all commercial banks soared during the latest trend period from the week of Feb. 22, 2023 to Mar. 22, 2023, with average increase of 162963.53 millions dollars per week. The upward trends were from Aug. 24, 2022 covered five trends.
Recent 7 trends of US deposits for all commercial banks with the slope of the trend.
The recent latest trend shows the U.S. banks are still under stress when the deposits continued to flee at an unprecedented rate of 40.2 billion dollars per week.
軟件：MathGPT for numerical modelling (lines combined method)
1. 在新冠肺炎初期最嚴重時，line 2顯示每週平均增長91 billion dollars。這是所有趨勢中最快增長銀行信貸趨勢的期間。
2. 從line 11到 line 16 持續每週平均增長超過20 billion dollars。
3. 直到line 17，美國所有商業銀行的銀行信貸隨著時間增長速度趨於緩和，並且從2022年12月7日當週開始，持續16週平均每週增長5 billion dollars。
1. ADP nonfarm payrolls had been growing slowly than before, seen average increase of 205.3 thousands per month. So the values in Green are less than ones in Orange.
2. The ADP nonfarm payrolls were lower than the values in Orange and in Green which are the trend level of payrolls without COVID-19 shock.
Orange: regress ADP nonfarm on time variable before 2020, and then keep the time code in COVID period to calculated the values.
Green: follow the line 10 from the results of setting of at least 12 months, and then keep the time code in COVID period to calculated the values.
2. #Covid19 hits the US employment market from the evidence of 15 months of line 11 and 23 months of line 12.
3. US employment market spent 15 months and had the average decrease of 402.8 thousands person who left their work per month . It spent 23 months climbing back to the employment level of 2019.
1. Before 2020, each trend covers 12 months by AI and data. Every year had stable and growing trend.
US Total Nonfarm Private Payroll Employment, Thousands of Person, Monthly, Not Seasonally Adjusted
Period: Jan. 2010 ~ Feb. 2023
MathGPT for numerical modelling (lines combined method) software
lines 23 and 25 show the reason why US employment market strongly growed.
line 26 shows why the economic experts and strategic analysts considered USA is entering or will enter the recession.