Median Sales Price of Houses Sold for the United States
- Rows
- 255
- Columns
- 2
- Data range
- 1963-01-01 to 2026-07-01
- Auto-refresh
- Re-checks its source every 14 days
- Last refreshed with new data
- 2026-10-29
Quarterly median price of new single-family houses sold in the US.
AI analysis
Dataset Analysis: Median Sales Price of Houses Sold (MSPUS)
1. Dataset Overview & Value for Correlation Studies This dataset captures the quarterly median sales price of new single-family homes in the United States, sourced directly from the Federal Reserve Bank of St. Louis's FRED API (https://fred.stlouisfed.org/graph/fredgraph.csv?id=MSPUS). FRED is one of the most authoritative and well-maintained macroeconomic data repositories in the world, and the source URL confirms this is a live API feed — meaning the dataset extends through Q1 2026, making it exceptionally fresh and suitable for near-real-time economic analysis. Spanning over 63 years of quarterly observations (1963–2026), this dataset is a cornerstone resource for housing market trend analysis, inflation studies, and long-run wealth inequality research. Its long temporal coverage makes it particularly powerful for identifying lagged correlations with macroeconomic indicators across multiple economic cycles.
2. Data Quality Observations Data quality is excellent across the board. With 253 rows and 2 columns, there are zero null cells in either column, meaning no observations are missing for any quarter in the covered range. The observationdate column has exactly 253 distinct values — perfectly matching total row count — which confirms no duplicate records and a clean, uninterrupted quarterly time series. The MSPUS price column similarly shows 231 distinct values out of 253 rows, which is entirely expected given that prices in some quarters may coincide, particularly in early decades when price movements were smaller in absolute dollar terms. There are no type mismatches reported, and the integer typing for housing prices is appropriate given that FRED reports these figures rounded to the nearest hundred dollars. This dataset requires no cleaning before use and is immediately join-ready.
3. Key Column Distributions & Statistical Insights The MSPUS column tells a compelling economic story through its statistics. The mean of $161,191 versus a median of $130,000 reveals a moderate right skew (skew = 0.68), meaning a smaller number of very high-price quarters in recent years pull the average above the typical historical observation — consistent with the post-2020 housing price surge. The wide interquartile range (Q1 = $59,000 → Q3 = $238,400, a spread of ~$179,400) reflects the dramatic multi-decade price appreciation across the full time series. The minimum of $17,800 (likely from the early 1960s) versus a maximum of $442,600 (likely 2022–2023 peak) represents a nominal 25x increase, though this conflates real appreciation with inflation. The standard deviation of $120,258 — nearly equal to the mean — underscores how non-stationary this series is, which is a critical consideration: correlations should generally be run on differenced or log-transformed values rather than raw levels to avoid spurious results.
4. Recommended Join Key for Cross-Dataset Correlation The observationdate column is the natural and only join key, and it is ideal for time-series merging. It is fully populated (0 nulls), perfectly unique (253 distinct values = 253 rows), and formatted as a standard ISO date (YYYY-MM-DD), which is broadly compatible across data platforms. Since observations are quarterly (Q1 anchored to January 1), any partner dataset should either be quarterly-native or require aggregation from monthly/annual frequency before joining. Analysts should verify that partner datasets use the same quarter-start convention (e.g., January 1 vs. March 31 labeling) to avoid a one-quarter offset error that is common when merging FRED data with BLS or Census sources.
5. Recommended Paired Datasets for Correlation Discovery Several dataset categories would pair exceptionally well with MSPUS for revealing economically meaningful correlations:
- 30-Year Fixed Mortgage Rate (FRED: MORTGAGE30US) — The most natural pairing; rising rates typically suppress prices with a lag of 2–4 quarters, making this ideal for Granger causality testing. - Consumer Price Index / CPI (FRED: CPIAUCSL) — Essential for deflating MSPUS into real terms; also useful for testing whether housing leads or lags general inflation. - Federal Funds Rate (FRED: FEDFUNDS) — Monetary policy transmission to housing prices is a well-studied mechanism and would support multi-decade regime analysis. - Housing Starts or New Residential Construction (FRED: HOUST) — Supply-side complement; testing whether construction activity leads price normalization after surges. - Median Household Income (Census/FRED: MEHOINUSA672N) — Critical for affordability ratio analysis and long-run wealth gap studies; the divergence between income and home price growth is a major policy question. - S&P 500 or equity market indices — For cross-asset correlation studies examining capital flows between equities and real estate during risk-on/risk-off cycles.
Columns
- observation_date (date)
- MSPUS (integer)