Stock Markets Analytics Zoomcamp 2025

Homework 2: Working with Data in Pandas Statistics

Distribution of scores and reported study time for this homework.

Submissions

83

Median total score

14

Average total score

12

Score distribution

All values are points.

Questions score

Min
4
Median
14.0
Max
14
Q1
10.0
Avg
11.5
Q3
14.0

Learning in public score

Min
-
Median
0.0
Max
3
Q1
0.0
Avg
0.2
Q3
0.0

Total score

Min
4
Median
14.0
Max
17
Q1
10.5
Avg
11.7
Q3
14.0

Time distribution

All values are hours reported by students.

Lectures

Min
0.0
Median
3.0
Max
16.0
Q1
2.0
Avg
3.4
Q3
4.0

Homework

Min
2.0
Median
5.0
Max
48.0
Q1
4.0
Avg
7.7
Q3
9.0

Question breakdown

Correctness and answer distribution per question.

1. What is the total withdrawn IPO value (in $ millions) for the company class with the highest total withdrawal value?

76 / 83 correct (91.6%)

1 2257 4 (4.8%)
2 4021 76 (91.6%)
3 3564 2 (2.4%)
4 6280 1 (1.2%)

2. What is the median Sharpe ratio (as of 6 June 2025) for companies that went public in the first 5 months of 2024?

66 / 83 correct (79.5%)

1 0.03 4 (4.8%)
2 0.14 7 (8.4%)
3 0.08 66 (79.5%)
4 0.20 4 (4.8%)

3. What is the optimal number of months (1 to 12) to hold a newly IPO'd stock in order to maximize average growth?

56 / 83 correct (67.5%)

1 10 10 (12.0%)
2 7 4 (4.8%)
3 4 10 (12.0%)
4 2 56 (67.5%)

4. What is the total profit (in $thousands) you would have earned by investing $1000 every time a stock was oversold (RSI < 25)?

70 / 83 correct (84.3%)

1 24 70 (84.3%)
2 30 3 (3.6%)
3 15 3 (3.6%)
4 42 2 (2.4%)

5. Q5. [Exploratory, Optional] Predicting a Positive-Return IPO

83 / 83 correct (100.0%)

Answer Count
Factor into the industry trends, revenue growth path, credibility of leadership and venture firm profiles 1
# Ans: Try to target IPOs with strong fundamentals (profitability, revenue growth, competitive edge). 1
no idea 1
HDB 1
Avoid "hot" IPO markets; buy during periods when pricing is softer. Focus on IPOs with strong financials and revenue growth before going public. Look for companies backed by reputable underwriters or with strong insider buying. Combine technical indicators with fundamental analysis. Use sentiment analysis on news and social media to gauge market perception. Track sector trends and invest in IPOs from growing or hot industries. 1
Given the recent sharp drops and recovery, I would use a smaller window, like 5 days. 1
AIRO AIRO Group Holdings, Inc 1
Invest only in fundamentally strong IPOs that show early positive momentum, are priced reasonably, and align with favorable sector and macro trends. 1
To increase the profitability of IPO investing, I would avoid buying on the first day and instead evaluate performance after the initial volatility settles. Additionally, filtering companies by profitability, sector momentum, and insider activity can help identify stronger candidates. Combining fundamental data with post-IPO technical indicators or even building a scoring model could further improve the selection process. 1
To improve profitability when investing in IPOs, delay entry until post-IPO stabilization, apply quality and technical filters (like RSI or momentum), avoid bear markets, and optionally use machine learning to predict positive-return candidates. 1
I will skip IPOs with extreme media hype or social media buzz—these often open high and underperform. 1
Instead of investing $1000 in every oversold IPO, implement a scoring system that filters for: Profitable companies with >20% revenue growth Low market volatility periods (VIX <25) Strong underwriters and institutional demand Sector momentum (avoid struggling sectors) Enhanced execution: Wait 2-4 weeks post-IPO for volatility to settle Dollar-cost average entries over multiple days Dynamic position sizing ($500-$2000 based on quality score) Stop-losses at -25% and profit-taking at +40% This transforms a spray-and-pray approach into a selective, risk-managed strategy targeting higher-quality opportunities. 1
To increase profitability, I would shift from a naive “buy-all” strategy to a selective strategy based on signals that historically correlate with positive post-IPO returns. Here are a few ideas: 1. Filter on Technical Indicators: Buy IPOs only when technical indicators (e.g. RSI < 25, MACD crossover) suggest a rebound or positive momentum. 2. Wait Period + Entry Timing: Avoid buying on IPO day. Instead, wait 10–20 trading days to let volatility settle and enter on a dip or trend reversal. 3. Fundamental Screening: Select IPOs with strong fundamentals (positive EBITDA, high revenue growth, low debt). 4. Macro Context Awareness: Invest only during favorable macro conditions (e.g. positive GDP growth, stable interest rates, bull market). 5. Sector Rotation: Prioritize sectors with recent momentum or favorable sentiment (e.g. tech in 2020, energy in 2022). 6. Avoid Overhyped IPOs: Exclude IPOs with excessive first-day jumps or extreme valuations, as they often underperform later. This would create a rule-based strategy that reduces downside exposure and focuses on quality setups. 1
* Predictive modelling: build a classification model to predict future returns using good feature engineering technique * Use news sentiment, web traffic, or social media hype metrics (e.g., Reddit mentions) to assess demand. * Use fundamentals if available: revenue growth, profitability (positive EBITDA), low debt-to-equity. * Combine low RSI (e.g., < 30) with volume surges or MACD crossovers for timing entries on pullbacks. * Use macro indicators (e.g., VIX, interest rate trends) to time entries. 1
Wait 30 days post-IPO, screen for positive momentum, strong fundamentals, and market tailwinds, then allocate with a fixed holding period or trailing stop. 1
Company Classification Analysis: Use company categorization (as in question 1) Focus on classes with better historical returns Optimal Entry Timing: Wait 2 months after IPO (as shown in question 3) Use technical indicators (RSI < 25) for entry Metric Filtering: IPO size (filter out very small ones) First-day trading volatility assessment Trading volume analysis Risk Management: Sector diversification Position size limits Stop-losses at 15-20% 1
Check a return in several days, e.g. 15 or 30 days, and add industry of the company. 1
I would include a sentiment score for the upcoming IPO, as it could provide insights into market expectations. I also believe that market liquidity plays an important role in shaping investor interest in the IPO. 1
## “Wait-and-Score” Strategy (Post-IPO Selection and Timing) > Most IPO investment strategies deliver negative average and median returns because they are often driven by hype, overvaluation, and early insider selling. To increase profitability, I would shift away from buying on IPO day and instead adopt a "wait-and-score" strategy. This strategy involves waiting 3 to 6 months after the IPO before considering an investment. This period allows the market to absorb early volatility, insider lock-up expirations, and initial earnings reports. Once this window has passed, I would screen IPOs based on a combination of fundamental and technical factors. Specifically, I would target companies that demonstrate strong fundamentals (such as profitability or a clear path to it), positive momentum (price trading above the IPO price and key moving averages), and signs of insider confidence (high insider ownership and limited selling). Additional filters could include positive earnings surprises, reasonable valuations compared to peers, and underwriters with strong reputations. Rather than betting on individual names, I would build a diversified basket of high-scoring IPOs and apply strict risk management — including position sizing and stop-loss rules — to control downside. By avoiding the initial hype phase and focusing on quality and momentum, this approach increases the likelihood of capturing the long-term winners among IPOs while avoiding the majority that underperform. 1
I have not considered ways of changing the strategies for investing in IPOs, but by carefully picking IPOs you might be able to improve the returns. For example, focusing on industries/sectors that have historically done well and looking at the underwriters as well as the venture capaitalist and private equity firms involved. Do they have a successful reputation too? 1
To increase IPO profitability, avoid hype-driven entries and focus on quality companies with strong fundamentals and reputable backers. Use data and timing (like post-lock-up dips) to guide entries, and manage risk with diversified or hedged strategies. 1
Of course. Here is a short English translation of the strategy: To improve profitability from investing in IPOs, the core idea is to shift from a broad approach to a selective, data-driven strategy. I would implement this using a machine learning model. The steps are: Build a Predictive Model: Train a classification model (e.g., Logistic Regression) on historical IPO data. The model's job is to learn the common characteristics of past IPOs that delivered positive returns. Identify Key Features: The model would use pre-IPO data for its predictions, such as: Company Fundamentals: Revenue, profit margins, growth rate. Deal Structure: IPO size, offer price, underwriter reputation. Market & Industry Trends: Sector popularity and overall market sentiment (e.g., VIX). Define the New Investment Strategy: For any new IPO, we first use the model to generate a probability score of its future success. The New Rule: Only invest in an IPO if its predicted probability of success is above a high threshold (e.g., >70%). This strategy filters out most of the historically poor-performing IPOs and aims to significantly increase profitability. 1
The best way to boost your IPO profits is to be smart about which companies you pick and when you buy and sell. Start by zeroing in on IPOs with solid fundamentals—think companies with strong revenue growth (say, over 20% year-over-year) and actual profits, not just hype. Next, time your buys using technical signals, like waiting for the stock to dip into oversold territory (RSI below 25) within the first 90 days after the IPO. That’s often when you can snag a bargain. Finally, be disciplined about selling—aim to cash out after about 4 months, when IPOs often hit a sweet spot, or if you see a quick 20% gain or the stock gets overbought (RSI above 70). This strategy lets you ride the early excitement of an IPO while dodging the longer-term slumps that many new stocks face. Compared to the options in Question 4 (24, 30, 15, 42), this approach could push your returns toward 30 or even higher by focusing on quality companies and sticking to a sharp, rule-based plan. 1
Brainstorming ideas for a positive‑return IPO strategy Wait for fundamentals — enter only after the first earnings report and go long if YoY revenue > +30 % and positive free‑cash‑flow margin. Momentum gate — require 60‑day relative strength ≥ +15 % vs Nasdaq‑100; exit when it turns negative. Insider alignment — filter for ≥ 40 % insider/VC ownership and no Form‑4 selling within the first 90 days. Macro overlay — trade only when VIX < 20 and the 10‑yr yield is flat‑to‑falling. Lock‑up arbitrage — avoid (or short) through the 180‑day lock‑up expiry, then buy the post‑lock‑up dip. Sector rotation — focus on sectors with ETF inflows > +2 σ (e.g., AI in 2023, renewables in 2020). Option‑skew confirmation — take trades only when first‑week call > put IV skew is positive. Composite rule example: Buy on the first earnings‑day close if sector ∈ {Tech, Health}, insider stake ≥ 40 %, RS ≥ +15 %, VIX < 20; sell after 7 months or on RS flip. Back‑test (2018–2024) lifts median 12‑m return from −24 % → +18 % and Sharpe from 0.03 → 0.37. 1
I would increase the dataset, but over a shorter timespan. The markets fluctuate with too many factors to count and to build an algorithm to predict every single possible event or scenario would be impossible. For example, covid impacted the markets in a huge way, but if we collected data from covid, I think it would negatively impact our inference capabilities. Training a deep learning model over a shorter period of time, for example, 1-3 years but with multiple datapoints every day might make its predictive capabilities more current. In addition to that, we could also increase our technical indicators so that our deep learning models have more features to work with. 1
To build a more profitable IPO investment strategy, I would propose identifying common patterns among companies that have shown positive returns with low volatility. This includes analyzing the context in which they were launched—such as whether the overall market was bullish or bearish, the country of listing, and the performance of their industry. By comparing new IPOs with similar recent ones and incorporating simple rules (like filtering for bullish markets and low-volatility sectors), it's possible to design a more robust strategy—even without relying on complex predictive models 1
To improve the profitability of IPO investments, I believe that, based solely on technical analysis, include the first days of the closing price movement of the shares and the volume observed. The performance of IPOs depends on many factors but I believe that an important one is the sector in which they are located, so first of all I would eliminate the sectors that usually reflect negative impacts in historical, at least I would remove for example the pharmaceutical sector, health, etc.. On the other hand, I would add a variable of changes in the first days to be able to define whether its trend movement is upward or downward and a relationship with the observed volume. In my experience IPOs tend to adjust price in the first few months and a trend is observed after the dates of share sales by insiders. The segment to which they belong usually goes in line with the price to be observed so it is important to outline a strategy based on companies in the same line of business. Translated with DeepL.com (free version) 1
I'd try a mix of the following strategy refinements, by order of preference 5, 6, 2, 1, 3, 4: 1. Selectivity: Avoid the Hype Focus on companies with strong fundamentals, not just buzz. Look for profitability or a clear path to profitability, rather than relying on speculative growth. Analyze sector trends—some industries consistently perform better post-IPO than others. 2. Timing: Avoid First-Day Trading Many IPOs see inflated valuations on day one due to retail and institutional demand. Instead of buying at the opening, wait for post-IPO corrections when the price stabilizes. 3. Lock-Up Expiration Strategy Many insiders and early investors are restricted from selling for about six months. After this period, a significant price drop often occurs as they cash out. Buying after the lock-up expiration can give you a more accurate market price. 4. Look at Pre-IPO Financials Companies must file S-1 forms with the SEC before going public. Look for consistent revenue growth, manageable debt, and a strong competitive edge. 5. Consider Alternative Entry Points SPAC mergers, direct listings, and secondary offerings might offer better risk-reward ratios than traditional IPOs. Options trading—using puts or covered calls—can hedge against downside risk. 6. Use Quantitative Screening Analyze IPO performance data to identify traits of successful IPOs. Look at past returns, volatility, and sector performance for patterns. 1
To increase IPO profitability, I’d propose: Filter for Strong Fundamentals: Select IPOs with high revenue growth (>20% YoY), positive EBITDA, and low debt-to-equity ratios (<0.5) to target financially sound companies. Technical Signals: Invest only when RSI is oversold (<30) or MACD shows a bullish crossover within 30 days post-IPO to time entries better. Sector Momentum: Prioritize IPOs in high-growth sectors (e.g., tech, biotech) with positive industry ETF returns (>5% over 3 months). Short Holding Period: Hold for 1–3 months (21–63 trading days) to capture early momentum, as Question 3 suggests optimal growth around 6 months. Market Conditions: Invest during bullish market phases (e.g., S&P 500 up >1% over 30 days) to avoid adverse conditions. These steps aim to pick high-potential IPOs and optimize entry/exit timing for positive returns. 1
There is a lot of volatility and uncertainty when IPOs initially go public. It is then prudent to wait post IPO seasoning, when prices stabilize, to see how the stock really performs. 1
low IPO pop (which is first trading day close minus IPO price) and popular sector. the IPO pop indicates IPO heap which investment bank intend to lower the valuation and popular sector without the heap indicates strong business model. 1
Buy IPOs with positive FCF & rev growth > 20% 1
On top of the indicators, we can also look at the following: 1. Due diligence on the fundamentals 2. Use market sentiments 3. Buy IPO based on rarity within given period (by market cap - large / small or by sector ) 4. Pick the up and coming industry instead of sunset 5. Generative AI using trend / book from a succesful stock influencer 6. Gather key people trades - i.e. nancy pelosi 7. Insider trading - LOL im joking 1
This IPO investment strategy involves only buying stocks that show a positive return three months after their IPO, filtering out early underperformers. A $1000 investment is then made at that 3-month mark, and its performance is tracked over the following 6, 9, and 12 months to evaluate long-term returns.The strategy’s historical results show mixed outcomes, with most eligible IPOs delivering negative or modest returns, though 2023 and parts of 2024 offered some positive average gains, highlighting the variability and risk even among initially strong performers. 1
To increase profitability: 1) Analyze first month after IPO and use only second one to trade. 2) Increase period of holding to 10 months if first two months were good. 1
Filter for quality companies: Invest only in IPOs from companies with strong revenue growth, solid financials, and a unique or defensible business model. Invest in IPOs with top-tier underwriters: Prioritize IPOs managed by reputable investment banks, as these tend to have better due diligence and market support. Focus on outperforming sectors or themes: Concentrate on IPOs within sectors or themes that are currently showing strong momentum, such as technology or renewable energy. Avoid IPOs with excessive hype or overvaluation: Be cautious of IPOs that are priced far above industry averages or have received an unusual amount of media hype. Consider only IPOs showing strong early trading support: Look for IPOs that demonstrate stable or rising prices and high trading volume in the days immediately after listing. Wait until after the lockup period to invest: Delay investing until after the IPO lockup period expires to avoid potential price drops from insider selling. Use market timing: Enter IPO trades primarily during bullish market phases, when overall investor sentiment and risk appetite are higher. Apply quantitative/machine learning models to predict winners: Utilize data-driven models to analyze historical IPO outcomes and identify patterns linked to positive returns. Monitor insider activity and institutional backing: Favor IPOs where insiders are buying shares or where there is significant participation from respected institutional investors. Implement risk management (e.g., stop-loss orders): Protect your capital by setting predefined exit points to limit losses on underperforming IPOs. Consider post-IPO dip opportunities instead of buying at debut: Wait for possible price corrections after the initial trading excitement before making a purchase. Avoid sectors with historically poor IPO performance: Exclude IPOs from sectors that have a track record of delivering weak post-IPO returns, such as low-quality biotech or certain cyclical industries. 1
To increase profitability when investing in IPOs, one should avoid buying on the first day and instead combine signals like a high Sharpe ratio (indicating strong risk-adjusted performance), favourable RSI levels (e.g., buy only when RSI < 25), and optimal holding periods (e.g., 2 months) based on historical average returns. Additionally, filtering out weaker company classes like Acquisition Corporations and focusing on IPOs in fundamentally strong sectors can improve outcomes. A data-driven approach—using historical IPO data, company metadata, and market indicators—can further enhance strategy performance by predicting which IPOs are likely to deliver positive future returns. 1
Top 3 Evidence-Based Strategies for Positive-Return IPOs Invest in IPOs Priced Below Filing Range Why?: Companies that price below their initial filing range (e.g., target $14-$16 but launch at $12) have +15% median 1-year returns vs. overpriced IPOs (Loughran & Ritter, 2002). Data Signal: python df_ipos['Discount_to_Range'] = (df_ipos['Final_Price'] - df_ipos['Range_Midpoint']) / df_ipos['Range_Midpoint'] good_ipos = df_ipos[df_ipos['Discount_to_Range'] < -0.10] # >10% below midpoint Target Small-Cap IPOs with High Institutional Ownership Why?: IPOs with market cap <$1B and >20% institutional ownership at launch show +22% 1-year returns (Journal of Finance, 2018). Screening Code: python good_ipos = df_ipos[ (df_ipos['Market_Cap'] < 1_000) & # $1B in millions (df_ipos['Inst_Ownership'] > 0.20) ] Avoid "Hot Market" IPOs Why?: IPOs launched during peak market cycles (e.g., >30 IPOs/month) underperform by -12% (SEC Research). Avoidance Rule: python bad_ipos = df_ipos[df_ipos['Market_Heat'] == 'High'] # High = >30 IPOs/month 1
The price to earnings ratio and other similar metrics like debt to equity would be useful. 1
Identify many indicators and regressors with predictive values and combine to make the best possible prediction 1
Use textual sentiment and language patterns around the IPO (from filings, headlines, and/or social media) as features in a machine learning model to predict post-IPO performance 1
Use machine learning? 1
Tracking social media sentiment (public opinion, analyst views, media tone) maybe be an early proxy for market expectations and future stock performance. I will scrape Reddit threads or X tweets mentioning the IPO company, score them and compute a weekly average sentiment. 1
I would say, probably using ml algorithms to be trained on past similar ipo data, and predict growth and future growth close to 3 months, or also do time based exits when certain profits are met 1
Potential IPO strategy improvements: 1. Company Classification Analysis: - Use company categorization (as in question 1) - Focus on classes with better historical returns 2. Optimal Entry Timing: - Wait 2 months after IPO (as shown in question 3) - Use technical indicators (RSI < 25) for entry 3. Metric Filtering: - IPO size (filter out very small ones) - First-day trading volatility assessment - Trading volume analysis 4. Risk Management: - Sector diversification - Position size limits - Stop-losses at 15-20% 5. Additional Factors: - Underwriter analysis (financial institutions, investment banks, that manage the IPO) - Market condition assessment - Lock-up period consideration 1
Implement a buy and hold strategy 1
[Exploratory, Optional] Predicting a Positive-Return IPO (1 point)* # To predict whether an IPO will yield a positive 1-year return, we can use a combination of price-based, company-level, technical, and macroeconomic features. Early performance indicators like first-day return, short-term volatility, and momentum (e.g., 30-day or 90-day growth) offer strong signals. Metadata such as the sector, IPO month, and exchange listing may capture structural or seasonal patterns. Technical indicators like RSI, MACD, and moving average crossovers help identify early price momentum or reversals. Finally, market conditions—such as S&P 500 trends, VIX levels, or interest rates—provide useful context about the broader economic environment at IPO time. # Classification problem with a Target: growth_252d > 1 1
Wait 3–6 months post-IPO to let volatility settle and look for breakouts above IPO price as a bullish signal. 1
Our analysis shows that peak returns occur within a 2-month period. Maybe we should test daily returns over a 1-60 day window to identify profitable opportunities. Another possible strategy could be shorting 1
To improve profitability, I would shift from passively buying all IPOs to an active strategy of selecting only fundamentally strong companies in promising sectors, while using technical indicators like RSI and moving averages to optimally time entries and exits. 1
Perhaps using different RSI thresholds to account for different volatility across sectors 1
Delay Entry – Wait for the Post-IPO Dip, wait 3–6 months after the IPO before buying the stock 1
Incorperating sentiment analysis on financial news can be a good direction to start with, as it is able to capture the market reactions (whether or not people are keen to buy), which can be very impactful towards an IPO. 1

Calculated: 25 June 2025, 06:51