samuel-da-shadrach

About:
Show more

No Scores Yet

Relative Brier Score

0

Forecasts

0

Upvotes
Forecasting Activity
Forecasting Calendar
No forecasts in the past 3 months
 

Past Week Past Month Past Year This Season All Time
Forecasts 0 0 37 0 37
Comments 0 0 4 0 4
Questions Forecasted 0 0 14 3 14
Upvotes on Comments By This User 0 0 6 0 6
 Definitions
New Prediction
samuel-da-shadrach
made their 2nd forecast (view all):
Probability
Answer
10% (+4%)
Less than $405 million
80% (+0%)
More than or equal to $405 million but less than $407 million
10% (-4%)
More than or equal to $407 million

Updating forecast because I saw good reasons in comments for budget to go down more likely than up (senate tends to allocate less), and I'm not sure whether they're weaker than my argument (original dept of science request larger). Just putting equal odds now in 1st and 3rd bin because I don't have much of an opinion.

Files
New Prediction
samuel-da-shadrach
made their 1st forecast (view all):
Probability
Answer
6%
Less than $405 million
80%
More than or equal to $405 million but less than $407 million
14%
More than or equal to $407 million

Informed by:

My own research on other Q on same bills:

https://www.infer-pub.com/comments/51029

@MullenAustin having done similar research on this Q:

https://www.infer-pub.com/comments/50857

Office of Science leadership

https://www.energy.gov/science/leadership

Director of Office of Science was appointed by Biden and seems notable aligned with Democrat issues such as climate, soil and women's rights. I'm not sure if this has any implication on whether Senate (republican majority) would be interested in increasing budget to the requested amount.

Deputy director Dr J Stephen Binks seems quite focussed on exascale computing and its implications for US national strategy. Talks of both AI and non-AI applications. This seems relatively non-partisan.

https://science.house.gov/imo/media/doc/Dr.%20Binkley%20Testimony.pdf


Ideally there should be an easy way to look at a lot of other bills passed by both appropriation committees, and see which committee's passed amounts have more weight. Does the resulting budget amount look like minimum (of house and senate app. committee recommendations) or average of the two or just what the house says or something else? Unfortunately I do not know any easy way to get this data, websites of both appropriation committees just have press releases which makes it difficult to compare.

https://appropriations.house.gov/news/press-releases

https://www.appropriations.senate.gov/news/majority

And it's hard to find the exact same bill on both sites. If someone knows a better way to get this data, feel free to reply my comment or ping me.


My forecast is therefore mostly just going to rely on what other people have said. Low likelihood of anything but the middle bin, and 3rd bin is more likely than the 1st bin because that's the dept of science requested in the first place.

Files
New Comment

Comment deleted on Aug 28, 2022 07:01AM UTC

Files
New Prediction
samuel-da-shadrach
made their 1st forecast (view all):
Probability
Answer
0%
$0
10%
More than $0 but less than $82 million
60%
Between $82 million and $87 million, inclusive
30%
More than $87 million

OUTSIDE VIEW

Most of my time went in looking for data

2022 enacted

$1,300M for innovation and improvement account

of which $234M for EIR

of which $82M for STEAM + CS

House bill for 2023

Committee recomends $1,484M for innovation and improvement account

Of which recommends $384M for EIR program

Of which recommends $87M for STEAM + CS

Senate bill for 2023

Recommends $1,361M for innovation and impr. acc.

Of which recommends $360M for EIR

Amount for STEAM + CS not specified

Find latest press releases by house appropriations committee here:

https://appropriations.house.gov/subcommittees/labor-health-and-human-services-education-and-related-agencies-117th-congress/congress_press_release

Signed by house appropriations committee in June 2022 (for FY23)

https://docs.house.gov/meetings/AP/AP00/20220630/114968/BILLS-117-FC-AP-FY2023-AP00-LHHS.pdf

$1,330M for innovation and impr.

Copy-pasting the breakdown provided in this report because I do not know how to decrypt this. See bottom of my comment [1]

What are these titles and subparts referring to, and how do they map to STEAM+CS budget? Here are the titles of the ESEA:

https://en.wikipedia.org/wiki/Elementary_and_Secondary_Education_Act#Title_II

Amounts specified: $194M, $652.5M, $384M, $8M, $100M

Signed by house appropriations committee in July 2021 (for FY22)

https://docs.house.gov/meetings/AP/AP00/20210715/113908/BILLS-117--AP--AP00-FY2022LHHSSubcommitteeAppropriationsBill.pdf

$1,297M for innovation and improvement

Breakdown not clear

Amounts specified: $300.5M, $642.8M, $254M, $100M

Could not find data (LHHE appropriations bill) for years prior to FY22 in the limited time I spent trying to find it. Older data would be useful for outside view forecast

INSIDE VIEW

TIL US House has democratic majority (220 dems 211 reps) whereas Senate has more republicans (48 dems 50 reps and 2 independents)

Possibly this means Senate recommendation will be slightly lower than house recommendation? Do not know

Spent some time trying to understand how committees and subcommittees work, not sure if I did.

Appropriation decisions need to be passed by both house and senate, hence final amounts will, I'm guessing, depend on whichever of the two pass a lower amount.

Will not have more time right now to dedicate to this question


Forecasting heavy weight on $82M to $87M inclusive, because that seems the range in which most figures are mentioned. The default if there is delay in passing of bill is also this bin, although this imo unlikely (<10% likelihood, but I haven't seen historical data for this either). Budgets tend to increase by default (if measured in dollars) due to inflation, hence putting some weight on $87M. Other reason for $87M is the optimistic house bill. There is no available reason for budget to go below $82M, only reason is something unknown.




[1] See copy-pasted stuff from appropriations-committee signed document.

"For carrying out activities authorized by subparts 1,

9 3 and 4 of part B of title II, and parts C, D, and E and

10 subparts 1 and 4 of part F of title IV of the ESEA,

11 $1,330,500,000: Provided, That $194,000,000 shall be for

12 subparts 1, 3 and 4 of part B of title II and shall be made

13 available without regard to sections 2201, 2231(b) and

14 2241: Provided further, That $652,500,000 shall be for

15 parts C, D, and E and subpart 4 of part F of title IV,

16 and shall be made available without regard to sections

17 4311, 4409(a), and 4601 of the ESEA: Provided further,

18 That section 4303(d)(3)(A)(i) shall not apply to the funds

19 available for part C of title IV: Provided further, That not20 withstanding section 4601(b), $384,000,000 shall be

21 available through December 31, 2023 for subpart 1 of

22 part F of title IV: Provided further, That of the funds

23 available for subpart 4 of part F of title I not less than

24 $8,000,000 shall be for continuation grants for eligible na25 tional nonprofit organizations, as described in the ApplicaVerDate Nov 24 2008 13:15 Jun 24, 2022 Jkt 000000 PO 00000 Frm 00149 Fmt 6652 Sfmt 6201 C:\USERS\APAUL\APPDATA\ROAMING\SOFTQUAD\XMETAL\11.0\GEN\C\L230624.XML

June 24, 2022 (1:15 p.m.)

H:\8-LABOR HHS ED\117TH CONG, 2ND SESSION\HOUSE SUBCOMMITTEE\BILL TURNS\XML\L230624.XML

L:\VA\062422\A062422.004.xml

150

1 tions for New Awards; Assistance for Arts Education Pro2 gram—Arts in Education National Program published in

3 the Federal Register on May 7, 2018, for activities de4 scribed under section 4642(a)(1)(C): Provided further,

5 That $100,000,000 shall be for competitive grants to local

6 educational agencies and State educational agencies to re7 duce racial and socioeconomic segregation across and

8 within school districts. "


Files
New Prediction

OUTSIDE VIEW

Hardly any protests or riots in Saudi Arabia over the past 4 years, as per ACLED. Hardly anay fatalities associated with protests or riots.


But using outside view for these kinds of long-tail driven events seems particularly bad, so inside view will matter a lot.


INSIDE VIEW

This Q is completely out of my depth as I have no prev. background in political affairs or Saudi politics, so I had to spend about 20 min reading about Saudi politics first. MbS' rule seems to have strengthened over time, although more authoritarian than his predecessors. This increase in centralised control seems like the main reason one might expect dissent to be worse for his regime than predecessors. However based on other regimes (not related to saudi arabia) (of which I know very little), I don't think this matters all that much, if anything, successful exertion of control can reduce odds of revolt.

Kidnapping of Lebanon PM was fun to read about (sorry), although I'm not sure if it has implications for this Q. Anyways it is over. I guess it is possible for some event in Iran-Saudi conflict to trigger tensions within Saudi Arabia itself. As always I feel out of my depth to understand how this link works. I will not have time to study much about all the various facets of Iran-Saudi proxy wars right now. Yemeni Civil war might be relevant, Saudi-led conflict has started pursuing peace talks in Mar 2022, it may be possible these fail.


I have no clue why this Q has picked the eastern province as the one to focus on. Eastern province is run by Nayef Al Saud, who is Nayef's older brother, Nayef being the crown prince who was deposed by MbS in 2017. However since Nayef Al Saud has successfully held positions of authority all the way till 2022, this does not seem like a major reason. It seems unlikely that members of the saud family would instigate civilian protests to resolve their disputes, although as always I don't have much of a clue, given that the dispute has previously escalated to the point of MbS overthrowint Nayef. The eastern province has noticeable border with Yemen, I wonder if this is a reason. Basically I don't really know.

I'm going with 10% because I have no real reason to expect protests, but have large uncertainty and hence don't wanna forecast more precisely than saying odds are less than 1 in 5. (Less than 1 in 5 would imply forecasting 10%)

Files
New Badge
samuel-da-shadrach
earned a new badge:

Upvotes Received

New Prediction
samuel-da-shadrach
made their 2nd forecast (view all):
Probability
Answer
50% (+3%)
Less than 20%
44% (-3%)
More than or equal to 20% but less than 24%
6% (+0%)
More than or equal to 24%

see reasoning in prev forecast

Files
New Prediction
samuel-da-shadrach
made their 1st forecast (view all):
Probability
Answer
47%
Less than 20%
47%
More than or equal to 20% but less than 24%
6%
More than or equal to 24%

OUTSIDE VIEW

Exported data to CSV, calculated percentage and plotted charts, performed linear regression

Over past 20 years, biofuels grow 120 T BTu / yr, total renewable grows 342 T BTu / yr. 

Projecting from 2021 data we get:

(2315 + 2 * 120) / (12156 + 2 * 342) = 2555 / 12840 = 0.199

Volatility could go either way, meaning 50% below 0.2 and 50% above 0.2

Outliers from total renewable trendline are up to 600 T Btu off.
Outliers from biofuel trendline are up to 400 T Btu off.
Outliers occur in 5 to 10 years
If you specify a direction then even rarer.

most optimistic outlier: (2555 + 400) / (12840 - 600) = 0.241
Barely makes it to 24%, and this involves 2 outliers at same time

Outside view forecast: 47% 47% 6%

INSIDE VIEW

Haven't spend enough time on this, but can't find inside view news on biofuel that affects within a single year. Main news is fossil fuel shortage due to Russia, but those don't directly impact this question. Russia does not export biofuel to US. I'm not sure if Russia is involved in any other renewable energy significantly. Russia exports wood to US, although don't have time to study how much impact this has on US wood fuel. Maybe next time I look at this question.

Also unsure of impact on fossil fuel prices going up, on renewable energy production and transport.

Will basically forecast the outside view for now.

Files
New Badge
samuel-da-shadrach
earned a new badge:

Active Forecaster

New Prediction
samuel-da-shadrach
made their 1st forecast (view all):
Probability
Answer
10%
In 3Q 2022 or 4Q 2022
10%
In 1Q 2023 or 2Q 2023
10%
In 3Q 2023 or 4Q 2023
70%
Not before 1Q 2024

OUTSIDE VIEW

Last mention was in Q1 2018, and 17 quarters have passed by without a  single mention. Assuming probability p of mention each quarter, then find probability for which our situation has atleast 50% probability. (1-p)^17 > 0.5 gives p < 0.04 per quarter. Which gives outside view forecast of:
< 0.04 + 0.96 * 0.04 = 0.0784
< 0.96^2 * 0.0784 = 0.0722
< 0.96^2 * 0.07225 = 0.0666
> 1 - (0.0784 + 0.07225 + 0.0666) = 0.7828

The proper way to do this is not assume P(17 quarters no mention) < 50% but to invert distribution or use laplace rule of succession or something but I haven't understood those methods yet.

INSIDE VIEW

Algal biofuels have not seen significant research advances since 2018 as far as I understand. 

However we may want to use slightly higher probabilities since:
a) the company gave a (weak) deadline of 2025
b) global interest in seeing alternative energy succeed increases with time, hence corporate boards may wish to emphasise them more in reports even if they don't have useful R&D results
c) Actual progress in algal biofuel might be non-zero since 2018 and if so we should expect it to continue.

I'm basically ignoring (b) because timeframe in which this pressure will increase is slow. I will slightly adjust probabilities upwards due to (a). (c) is hard to judge without deep view, I couldn't figure this out just by skimming abstracts of few papers.

Files
Files
Tip: Mention someone by typing @username