◆ Statistics

What an econometrician
really does.

20 tasks, each one witnessed by the sources that watched the job — and behind every one, a prompt you can use tonight.

20evidenced tasks
29,030in the US (2025)
$105,650median pay / year
5systems it runs on
This is what one task looks like here
Apply statistical methods to data
Estimate treatment effects using the cleaned panel, run difference-in-…4 sources agree

The shape of the day

tap a movement to see its tasks

Which one is you, right now?

Pick the moment · no score, no sign-up
Which moment is you right now?
Whichever you pick, the task behind it opens below.

The work, task by task

20 tasks
Hands on the work16
Apply statistical methods to data+
Estimate treatment effects using the cleaned panel, run difference-in-differences with clustered standard errors, test parallel trends, and produce a short methods note with robustness checks by Wednesday.
escojdonetwiki4 agree
when the reply comes backPush once: ask it to sharpen the weakest part, and to say what it assumed. Helpful?
Analyze data to identify trends+
Produce time series and cross-section descriptions to detect emerging trends in consumption and employment, run seasonal decomposition and moving averages, flag three candidate trend stories and visualise each by Thursday.
jdonetwiki3 agree
when the reply comes backPush once: ask it to sharpen the weakest part, and to say what it assumed. Helpful?
Produce statistical reports and visualizations+
Produce a six-page statistical report with tables and three visualisations summarising income distribution, unemployment rates, and treatment heterogeneity, include methodology appendix and deliver PDF by next Tuesday.
jdonetwiki3 agree
when the reply comes backPush once: ask it to sharpen the weakest part, and to say what it assumed. Helpful?
Identify statistical patterns+
Scan the time-series and cross-sectional datasets for nonstationarity, structural breaks, and correlated residuals, flag candidate explanatory variables with clear partial correlations, and produce a one-page memo summarising the patterns and proposed tests by Thursday.
esco
when the reply comes backPush once: ask it to sharpen the weakest part, and to say what it assumed. Helpful?
Gather data+
Compile the consumer transactions, macro indicators, and firm-level balance sheets into one cleaned panel, reconcile mismatched identifiers and missing dates, document data provenance and quality issues, and deliver the harmonised dataset and a 200-word provenance note by Tuesday morning.
esco
when the reply comes backPush once: ask it to sharpen the weakest part, and to say what it assumed. Helpful?
Evaluate and describe data utility+
Evaluate the utility of the household survey dataset for policy modelling: report variable coverage, temporal and geographic granularity, known biases, effective sample sizes per subgroup, and a clear recommendation on whether it can support causal inference.
jdwiki2 agree
when the reply comes backPush once: ask it to sharpen the weakest part, and to say what it assumed. Helpful?
Use software like R, Python, SAS, SQL+
Prepare an analysis workflow using R and SQL: list the extraction queries, cleaning steps, feature transformations, and statistical packages to use, plus expected runtimes and one reproducible script that pulls a test sample and returns variable summaries.
jdwiki2 agree
when the reply comes backPush once: ask it to sharpen the weakest part, and to say what it assumed. Helpful?
Build models and perform hypothesis testing+
Build the econometric specification, estimate the models for treatment and control groups, run robustness checks and standard hypothesis tests, then write a results appendix with coefficients, standard errors, p‑values and interpretation for the policy team.
jd
when the reply comes backPush once: ask it to sharpen the weakest part, and to say what it assumed. Helpful?
Watch and assess3
Grow the practice1

What the work runs on

named inside the evidenced tasks
10 tasksMicrosoft Excelprepare cleaned input tables and summary statistics for the noteOpen its task library →
7 tasksIBM SPSS Statisticsimplement regression models, clustered SEs, and standard diagnostic tests for panel data
5 tasksMicrosoft Officeassemble the report, format tables and export the final PDF
2 tasksApache Sparkprocess large panel datasets efficiently and compute time-series decompositions at scale
1 taskLinuxuseful for scripting batch imports and managing file organisation when datasets are large or automated

The same task, four heights

this page is height one
ExecuteDo today's task, with fewer mistakesyou are here → ImproveMake it easy for the next person to acceptin the atlas → DecideWork out the right move when it is unclearin the atlas → BecomeLearn the pattern so it stops coming backin the atlas →

Can AI actually do this job?

the honest answer

It can

where it genuinely helps
  • Explain the theory behind the work
  • Draft, tidy and structure your writing
  • Rehearse a hard conversation before you have it
  • Build a study plan that fits your gaps

It cannot

where it stops, completely
  • Be in the room where an econometrician actually works
  • Carry the responsibility when the call is wrong — that weight stays yours
  • Notice what no one wrote down: the hesitation, the thing left unsaid
  • Live with the outcome
M · R6 of R7This job sits on estimating — real theory, real diagnosis, but reality still holds the grading pen.

What the work pays

two countries, two different measures

United States

this exact occupation · BLS 2025
  • $105,650 a year — the middle: half earn more, half earn less
  • The lowest tenth earn near $64,000; the top tenth near $174,050
  • 29,030 people employed in this occupation

India

the occupation GROUP, not this job · PLFS via ILOSTAT 2025
  • ₹26,152 a month — the median for Technicians and associate professionals, the group this work sits in
  • India publishes pay by broad occupation group, so this covers many jobs besides this one. It is a shape, not a salary.
read this carefullyThese two numbers are not comparable and must not be converted into each other. One is a yearly figure for this job alone; the other is a monthly figure for a whole family of jobs. What travels between them is the pattern, not the amount: experience lifts pay almost everywhere.

Where the evidence lives

open any of it yourself

Close to this work

12 nearby
StatisticsForecasting Analyst26 evidenced tasks StatisticsExperimentation Scientist26 evidenced tasks StatisticsProcess Improvement Analyst25 evidenced tasks km/h RPMStatisticsMeasurement Analyst25 evidenced tasks StatisticsData Labeling Lead23 evidenced tasks StatisticsFreelance Data Consultant23 evidenced tasks StatisticsSas Programmer22 evidenced tasks StatisticsActuary21 evidenced tasks

Questions people actually ask

You spend most of the day on data: collecting raw files, cleaning them, and merging tables so they’re usable. That often means writing SQL queries, running Python or R scripts, and checking data quality.

Afternoons commonly go to modelling and meetings. You build regression or time-series models, run hypothesis tests, then explain results to managers in plain charts made in Excel, R, or SPSS. Expect one or two meetings to set data needs or review results.

Start with one programming tool and one analytics package. Learn Python (pandas, statsmodels) or R for modeling and visualization, and SQL for extracting data from databases. These cover most daily tasks.

Also become comfortable in Excel for quick checks and presentations, and know how to run jobs on Linux. Knowing SAS or IBM SPSS Statistics is helpful in firms that use them, and Apache Spark is useful if you’ll handle very large datasets.

Use the Bureau of Labor Statistics (BLS) numbers for a reliable picture. BLS reports about 29,030 employed econometricians and a median salary of $105,650 per year. The lowest tenth earn about $64,000, while the top tenth make around $174,050.

Those numbers vary by industry, location, and experience. Government or academic roles may start lower; finance and tech firms often pay toward the higher end.

Treat AI/ML as tools for pattern finding, not final answers. Use algorithms in Python or Spark to explore patterns, but always run statistical tests and diagnostic checks (like residual analysis) so models aren’t just picking up noise.

Document data sources, test models on holdout data, and show uncertainty (confidence intervals, p-values). When models affect people, keep a human in the loop: managers must review high-impact decisions and you should explain assumptions plainly.

Econometrics focuses on causal inference and economic theory: you test hypotheses about cause and effect using regressions, instrumental variables, and time-series tools. It’s more about answering ‘why’ than just predicting.

Data science often emphasizes prediction and machine learning at scale (Spark, big feature engineering). Econometric work still uses Python/R and SQL, but you’ll spend more time on statistical principles and careful identification strategies.

Communication: turning statistical results into simple, actionable statements for managers. You can build the best model, but if you can’t explain assumptions, uncertainty, and limitations, it won’t be used.

Closely tied is data hygiene—knowing how to assess source reliability and clean messy inputs. Many projects fail because the data were wrong, not because the model was bad.

km/h RPM

How far can AI take you?

one question
Be honest — how far can a language model take you on estimating?
There is a real answer, and it is not the flattering one.