Class record · ADEC · ADV

ADEC 7430

Big Data Econometrics · 6 instructors on record · 23 sections

What the course score is made of

87BC evaluations23 records · weight ×1.0 · Spring 1 2023 – Spring 2026

—Student reviewsno reviews — does not contribute

—RateMyProfessorsno ratings — does not contribute

That is the class itself, averaged across everyone who taught it. Who taught it, and how each of them scored, is the board below — that is what a student picking a section is actually choosing between.

What the challenge rank is made of

82ndMore challenging than 82% of BC classesranked against every one we hold

96BC — “Intellectually Challenging”23 records · raw score out of 100

Boston College's own “intellectually challenging” question for this class only — students answer it positively about courses that stretched them, so it measures intellectual challenge and NOT grading, workload or quality. Almost every BC class scores high on it, which is why it is shown as a rank rather than a score.

What the workload rank is made of

94thMore work than 94% of BC classesranked against every one we hold

84BC — “Workload”23 records · raw score out of 100

Boston College reports workload for this class only as one of five bands, not as a number of hours, and it does not publish what the bands mean — so this site will not translate it into hours. Higher means more work, which is not the same as worse.

This course demonstrates how to merge economic data analysis and applied econometric tools with the most common machine learning techniques, as the rapid advancement of computational methods provides unprecedented opportunities for understanding "big data." This course will provide a hands-on experience with the terminology, technology, and methodologies behind machine learning with economic applications in marketing, finance, healthcare, and other areas. The main topics covered in this course include: advanced regression techniques, resampling methods, model selection and regularization, classification models (logistic regression, Naïve Bayes, discriminant analysis, k-nearest neighbors, neural networks), tree-based methods, support vector machines, and unsupervised learning (principal components analysis and clustering). Students will apply both supervised and unsupervised machine learning techniques to solve various economics-related problems with real-world data sets. No prior experience with R or Python is necessary.

The board

6 instructors · ranked on this class onlyCompare the top 3

Scores are real averages. The order also accounts for how many records back each one — the count is under every score — so one perfect review does not outrank a semester of them. 2 of the 6 rows carry a mark saying so.

Rank	Instructor	Overall	Workload	Challenge	RMP overall
1	Arvind SharmaAdvancing Studies	985 records	93rd	100th	—no page
2	Haydar EvrenEconomics Provisional fewer records	1001 record	78th	100th	—no page
3	Stefano ParravanoAdvancing Studies Provisional	902 records	100th	48th	—no page
4	Razvan VelicheAdvancing Studies	869 records	94th	66th	—no page
5	Nathaniel BastianAdvancing Studies Provisional fewer records	661 record	—	39th	201 rating · all classes
6	Ashuraj SirohiAdvancing Studies	775 records	93rd	89th	—no page

How this board is orderedPositions account for how many records stand behind each number, so one perfect review does not outrank a semester of them. Every score shown is the real average — only the ordering is adjusted. The comparison here is the class itself: a thin record is read against how this class was rated overall, not against the university.

Every instructor with an evaluation record for this class is on this board — it is not a top ten, and nothing is hidden behind a “show more”. Overall is the instructor's blended score for THIS class only, and its one contributing source is Boston College's evaluations for the sections in the row's record count — nothing on this board is an average across the other things they teach. Workload and Challenge are RANKS against every class we hold, not scores: BC's questions all sit near the top of the 1-5 scale, so a raw number would put almost every row in the high eighties and tell you nothing. Challenge is BC's “intellectually challenging” item, which students answer positively about courses that stretched them — it is not a measure of difficulty or grading. Neither column is ever marked better or worse. 3 of 6 rows are marked provisional: fewer than 3 records stand behind the number. That is thin evidence, not a bad instructor — the record count is in every row. The RateMyProfessors column is an instructor's overall figure across everything they teach; that site has no course dimension, so it is shown for information and kept out of the ranking.

The line

All 13 items · means across 23 records

The Professor4 items

Clearly Explains MaterialThe instructor was effective at creating an understanding of difficult subject matter.

8820th pctl

Available Outside of ClassThe instructor was available for help outside of class.

9029th pctl

Brings Out Your Best WorkThe instructor motivated me to do my best work.

8926th pctl

Sparked My InterestThe course stimulated my interest in the subject matter.

9129th pctl

The Course3 items

Well-Organized CourseThe course was well organized.

8937th pctl

Clear Learning GoalsThe learning objectives for the course were clear.

9146th pctl

Useful Beyond the CourseWhat I learned is applicable beyond this course.

9462nd pctl

Assignments & Feedback3 items

Assignments Help You LearnAssignments and exams contributed to my learning.

9254th pctl

Gives Feedback PromptlyI received feedback on my work in a timely manner.

8834th pctl

Meaningful FeedbackThe feedback I received on my work was meaningful.

9043rd pctl

Workload, difficulty & attendanceNot ratings

WorkloadHow much work the course took outside class, on Boston College's five-band scale — the band boundaries are not published, so this is not a number of hours.

8494th pctl

Intellectually ChallengingThe course was intellectually challenging.

9682nd pctl

Attendance MattersAttending class was necessary to do well in this course.

8417th pctl

Bar length is the rank against every comparable record we hold, not the raw score: BC's questions all have a median around 4.5, so a raw-scale bar would be nearly full on every row and would tell you nothing. Every bar is one ink colour — nothing here is graded green or red. The hatched bars are not ratings at all: a higher number on either describes a different course, not a better one.

Game log

23 sections · newest first

Instructor	Term	Sec	Instr	Course	Work
Arvind Sharma	Spring 2026	01	100	96	—
Arvind Sharma	Spring 2026	02	100	95	60
Arvind Sharma	Fall 2025	01	100	100	100
Razvan Veliche	Summer 2025	01	92	88	80
Stefano Parravano	Summer 2025	02	100	100	100
Arvind Sharma	Spring 2025	01	100	100	—
Razvan Veliche	Spring 2025	02	83	77	80
Arvind Sharma	Fall 2024	01	88	92	—
Razvan Veliche	Summer 2024	01	95	85	100
Razvan Veliche	Spring 2024	01	86	80	80
Razvan Veliche	Fall 2023	01	80	80	80
Razvan Veliche	Summer 2023	01	100	100	—
Stefano Parravano	Summer 2023	02	80	80	—
Ashuraj Sirohi	Spring 2023	01	80	85	—
Ashuraj Sirohi	Fall 2022	01	76	80	80
Haydar Evren	Summer 2022	01	100	100	60
Ashuraj Sirohi	Spring 2022	01	90	87	—
Ashuraj Sirohi	Fall 2021	01	58	65	80
Ashuraj Sirohi	Midterm Fall 2023	02	80	80	80
Nathaniel Bastian	Midterm Fall 2021	02	66	87	—
Razvan Veliche	Midterm Fall 2022	02	72	72	100
Razvan Veliche	Midterm Spring 2022	02	100	100	100
Razvan Veliche	Spring 1 2023	02	63	71	—

How to read this boardInstructors are ranked on their record in this class only, never on their average across everything they teach. Scores are out of 100 — Boston College's 1–5 agreement scale, ×20, the way these numbers were printed for years. Workload is BC's own five-band figure, not a number of hours, and is shown as a rank. A dash means the question was not reported, and is never a zero.