#NewData
LightNews LightNews
solomonkurz.bsky.social
Is there a nice way to do this with SPSS?

library(marginaleffects)

fit <- lm(data = mtcars, mpg ~ disp)

predictions(fit,
newdata = datagrid(disp = c(100, 200, 300)),
hypothesis = difference ~ pairwise)
March 16, 2026 at 3:58 PM
dingdingpeng.the100.ci
marginaleffects also supports newdata= (plus datagrid to create all sorts of counterfactual data grids), but variable= is a bit more convenient for standard use cases, I think.
November 30, 2025 at 4:32 PM
vincentab.bsky.social
Exactly! The `newdata` argument behaves exactly as in base R; `variables` is a new shortcut for G-computation.

avg_predictions(variables="D"): Marginal means w/ counterfactual values of D.

avg_comparisons(variables="D"): ATE of D, i.e. difference between "counterfactual" marginal means
November 30, 2025 at 7:10 PM
f2harrell.bsky.social
The classic R (and possibly preferred) way of stating this is newdata= instead of variable=.
November 30, 2025 at 4:28 PM
weare.rladies.org
broom::augment(model, newdata=...) – add model predictions or residuals to your dataset. If you pass the original data (or newdata), augment returns the data w/ extra columns like .fitted (predicted values) and .resid (residuals) - perfect for checking where the model may be off.

(5/n)
April 4, 2025 at 8:19 PM
konsta.happonen.eu
This is cool. @tslumley.bsky.social implemented a predict method for svyVGAM::svy_vglm models, which means that @vincentab.bsky.social's marginaleffects can now in theory be extended to accomodate these models.

#rstats
library(tidyverse) library(survey) library(srvyr) library(svyVGAM) library(marginaleffects) dd <- tibble( y = sample(LETTERS[1:5], 100, replace = TRUE), x = rnorm(100), weights = 1 + abs(rnorm(100, 0, 0.1)) ) dd <- dd |> as_survey_design(weights = weights) m1 <- svy_vglm(y ~ x, multinomial(refLevel = "A", ynames = TRUE), dd) source("R/marginaleffects_extension.R") m1 |> avg_comparisons( newdata = datagrid(x = 0), variables = "x", type = "response" )
April 28, 2026 at 2:32 PM
icpsr.bsky.social
Pumpkin spice, cozy reads, and fresh ICPSR data—our favorite fall trio! 🍂☕📚

Get fueled for your next research project by exploring the latest data releases. Your perfect dataset might be waiting! myumi.ch/ICPSR-new-releases

#ICPSR #NewData #DataDiscovery #Research
Hand holding a pumpkin spice latte next to an open book with a dried maple leaf on a cozy brown blanket. myumi.ch/ICPSR-new-releases
September 21, 2026 at 9:05 PM
jonminton.bsky.social
In particular this code pattern:

predictors_x0 <- data.frame(supp = "OJ", dose = 1.25)
predictors_x1 <- data.frame(supp = "VC", dose = 1.25)
predictors_fd <- rbind(predictors_x0, predictors_x1)

predictions_fd <- rstanarm::posterior_predict(
best_model_bayesian,
newdata = predictors_fd
)
January 25, 2025 at 1:29 PM
tinahinz.bsky.social
🚨New Data Relase🚨
Der Befragungsdatensatz zu TransFAR 2020 ist veröffentlicht!

Check: bamf.de/fdz

#MigrationResearch #newdata
October 27, 2025 at 3:49 PM
tinahinz.bsky.social
🚨New Data Release 🚨

Wir haben unsere Daten zu Berufssprachkursen aktualisiert und erweitert.
Wer zu Spracherwerb, Lernen in Gruppen und/oder Migranten forscht, sollte hier mal reinschauen!

Wir freuen uns auf eure Forschungsideen!

#newdata #migrationresearch
bamf.de Bundesamt für Migration und Flüchtlinge @bamf.de · Mar 17
📢 Das Forschungsdatenzentrum des BAMF stellt der Wissenschaftsgemeinschaft neue Forschungsdatensätze zu Berufssprachkursen (BSK) mit erweiterten Analysemöglichkeiten für die Jahre 2016 bis 2023 zur Verfügung.

Mehr dazu finden Sie unter www.bamf.de/fdz
Die Grafik zeigt das Foto eines jungen Mannes in einem hellen Büroraum mit Computern. Er schaut konzentriert auf den Laptop vor sich. Überschrieben ist das Bild mit: „Neue Daten des BAMF-FDZ zu Berufssprachkursen“.
March 17, 2026 at 12:30 PM
mtobis.bsky.social
I just completed "Red-Nosed Reports" - Day 2 - Advent of Code 2024 #AdventOfCode adventofcode.com/2024/day/2

A little bit more complicated today. It would be very tedious in C or similar. Necessary to create many lists similar to existing list with one element removed.
for index in range(len(data)): newdata = data.copy() del newdata[index]
December 3, 2024 at 3:08 AM
willsmithvision.bsky.social
Not only that, but LaTeX has elegant and intuitive programming syntax. For example, creating a new Value object only uses the word "expand" four times:
\expanded{
\noexpand\pgfoonew\expandafter\noexpand
\csname #2\endcsname=
new Value(\newdata,{\selfid,\otherid},*,0)
}
April 1, 2025 at 12:30 PM
wastewaterscan.bsky.social
By expanding national wastewater surveillance, these #data support infectious disease #epidemiology, early warning, and public health decision-making at scale. 📊

🔗 Read the article: www.sciencedirect.com/science/arti...

#WastewaterSurveillance #NewData
Pathogen nucleic acids data in wastewater solids from 147 treatment plants in the United States: 2024-2025
This data article provides human pathogen nucleic-acid concentrations in wastewater solids from 147 treatment plants across 40 states in the United St…
www.sciencedirect.com
January 29, 2026 at 2:13 PM
israelsenm.bsky.social
Our newdata show 38% of individuals with SLD are reclassified over 2 years ‼️

Over 60% with MetALD/ALD and 38% with MASLD change subclass.

This highlights that alcohol use and metabolic risk need regular reassessment, incl. for clinical trial eligibility 🧐

Link: www.cghjournal.org/article/S154...
Steatotic liver disease classification is dynamic, affecting clinical trial eligibility and subclass-specific treatments
Steatotic liver disease (SLD) includes the subclasses metabolic-dysfunction associated steatotic liver disease (MASLD), metabolic and alcohol-related liver disease (MetALD) and alcohol-related liver d...
www.cghjournal.org
April 8, 2025 at 6:13 AM
tjmahr.com
had to try it
plot of a jumpy quadratic curve m <- lm(prop_correct ~ poly(age_months, 2) + factor(age_years), data = okay) broom::augment(m, newdata = okay) |> ggplot() + aes(x = age_months, y = .fitted) + geom_point() + labs(title = "prop_correct ~ poly(age_months, 2) + factor(age_years)")
July 16, 2025 at 4:12 PM
jampants.net
error: subscripted value is neither array nor pointer\n newData[i] = data[i];\

yay...
February 21, 2025 at 3:34 AM
icpsr.bsky.social
On the lookout for quality data? 👀 Look no further than ICPSR! Check out this week’s new data releases and updates at myumi.ch/ICPSR-new-releases + sign up for weekly updates straight to your inbox: myumi.ch/5yzNV

#Data #NewData #DataUpdates #Research #DiscoverData #ICPSR
Meerkat standing alert on a rock next to text promoting ICPSR's weekly data updates and sign-up links.
June 23, 2026 at 2:30 PM
stephenjwild.bsky.social
My only suggestion is entering the data in newdata, so you have

avg_slopes(mod, newdata = d_freechoice).

Sometimes the R environment can lead weird errors if you don't.
November 30, 2023 at 2:38 PM
vincentab.bsky.social
Issue is now closed with the inconsistency resolved in the `main` dev branch on Github.

No, the difference in numeric values is due to the fact that when using `model=`, `insight::get_data()` calls `na.omit()`, whereas we take the `mean()` of the whole dataset with `newdata=dat`
September 1, 2025 at 3:24 PM
zxvintage40.bsky.social
Screen Shower for Mission Ninety One/Newdata Software (UK) (1986). #zxspectrum #retrogaming

WATCH ON YOUTUBE: https://www.youtube.com/watch?v=lrAXSkYMyV4
February 4, 2026 at 8:40 PM
wat3rm370n.bsky.social
Here's a link to the PA DCNR map tool you can use to measure distance or area. I seem to use this most around data center issues anymore. I'd really like to get back to the days I used this just to carefully plan hiking and kayaking trip itinerary!

newdata-dcnr.opendata.arcgis.com/apps/DCNR::p...
newdata-dcnr.opendata.arcgis.com
April 21, 2026 at 2:01 AM
icpsr.bsky.social
📈 We’re thrilled to share the update of the study “Synthetic Data Generation of Health and Demographic Surveillance Systems Dataset, Kenya, 2019-2020 (ICPSR 39209)”.

👉 Explore the data: myumi.ch/n167A

#ICPSR #NewData @umichmedicalschool.bsky.social
New Data Release: Synthetic Data Generation of Health and Demographic Surveillance Systems Dataset, Kenya, 2019-2020 (ICPSR 39209). Image of the Kilifi county of Kenya flag on a flagpole waving in the bright sunrise with mist and fog.
August 25, 2025 at 2:19 PM
paolo.ricciuti.me
You can also do this

```
<script>
let { data } = $props();
let newData = $derived.by(()=>{
let _ = $state(data);
return _;
});
</script>
```
And just modify newData.stuff
December 17, 2025 at 11:26 AM
vincentab.bsky.social
In addition, there is a slight inconsistency in behavior that I should fix. See here: github.com/vincentarelb...
`datagrid()` inconsistency with `model` and `newdata` arguments · Issue #1584 · vincentarelbundock/marginaleffects
When supplying model argument, datagrid() properly detects binary variables as "binary", whereas it treats them as integers when supplying only the newdata argument. This has implications in some c...
github.com
September 1, 2025 at 3:06 PM
s3alfisc.bsky.social
We’ve released a patch to pyfixest (0.26.2) with the following new features:

- feols and fepois have new split and fsplit arguments (just as in fixest!)
- you can check for separation in Poisson models via the „iterative rectifier“
- we’ve fixed a 🐞 in the predict method for WLS with newdata
November 1, 2024 at 11:58 AM