Aggregation Functions

QuestDB provides a comprehensive set of aggregate functions optimized for columnar storage and SIMD acceleration.

Basic Aggregates

COUNT

Count rows or non-NULL values:

SELECT count() FROM trades;
SELECT count(price) FROM trades;

Query Result:

count
10000

Count distinct values:

SELECT count_distinct(symbol) FROM trades;

SUM

Sum numeric values:

SELECT sum(quantity) AS total_volume
FROM trades;

Query Result:

total_volume
1250000

AVG

Calculate average:

SELECT 
    symbol,
    avg(price) AS avg_price
FROM trades
GROUP BY symbol;

Query Result:

symbol  avg_price
AAPL    150.75
GOOG    142.50
MSFT    250.25

MIN / MAX

Find minimum and maximum values:

SELECT 
    symbol,
    min(price) AS low,
    max(price) AS high
FROM trades
GROUP BY symbol;

Query Result:

symbol  low     high
AAPL    148.20  152.80
GOOG    140.00  145.00
MSFT    248.50  252.00

Statistical Aggregates

STDDEV / STDDEV_SAMP

Sample standard deviation:

SELECT 
    symbol,
    stddev_samp(price) AS price_volatility
FROM trades
GROUP BY symbol;

STDDEV_POP

Population standard deviation:

SELECT 
    symbol,
    stddev_pop(price) AS pop_stddev
FROM trades
GROUP BY symbol;

VAR_SAMP

Sample variance:

SELECT 
    symbol,
    var_samp(price) AS price_variance
FROM trades
GROUP BY symbol;

VAR_POP

Population variance:

SELECT 
    symbol,
    var_pop(price) AS pop_variance
FROM trades
GROUP BY symbol;

COVAR_SAMP

Sample covariance:

SELECT 
    covar_samp(price, volume) AS price_volume_covariance
FROM trades;

COVAR_POP

Population covariance:

SELECT 
    covar_pop(price, volume) AS pop_covariance
FROM trades;

CORR

Correlation coefficient:

SELECT 
    corr(price, volume) AS price_volume_correlation
FROM trades;

Positional Aggregates

FIRST

Return first value:

SELECT 
    symbol,
    first(price) AS opening_price
FROM trades
GROUP BY symbol;

LAST

Return last value:

SELECT 
    symbol,
    last(price) AS closing_price
FROM trades
GROUP BY symbol;

FIRST_NOT_NULL

Return first non-NULL value:

SELECT 
    symbol,
    first_not_null(price) AS first_valid_price
FROM trades
GROUP BY symbol;

LAST_NOT_NULL

Return last non-NULL value:

SELECT 
    symbol,
    last_not_null(price) AS last_valid_price
FROM trades
GROUP BY symbol;

String Aggregates

STRING_AGG

Concatenate strings:

SELECT 
    date_trunc('day', timestamp) AS day,
    string_agg(symbol, ',') AS symbols_traded
FROM trades
GROUP BY day;

Query Result:

day                          symbols_traded
2023-12-28T00:00:00.000000Z  AAPL,GOOG,MSFT,AAPL,GOOG

Percentile Aggregates

PERCENTILE_APPROX

Approximate percentile (faster):

SELECT 
    symbol,
    percentile_approx(price, 0.5) AS median,
    percentile_approx(price, 0.95) AS p95
FROM trades
GROUP BY symbol;

Query Result:

symbol  median  p95
AAPL    150.50  152.00
GOOG    142.00  144.50

GROUP BY

Group rows for aggregation:

Single Column

SELECT 
    symbol,
    count() AS trade_count,
    sum(quantity) AS total_volume,
    avg(price) AS avg_price
FROM trades
GROUP BY symbol;

Multiple Columns

SELECT 
    symbol,
    date_trunc('day', timestamp) AS day,
    count() AS trade_count,
    sum(quantity) AS volume
FROM trades
GROUP BY symbol, day
ORDER BY day DESC, symbol;

Query Result:

symbol  day                          trade_count  volume
AAPL    2023-12-28T00:00:00.000000Z  250          25000
GOOG    2023-12-28T00:00:00.000000Z  180          18000
MSFT    2023-12-28T00:00:00.000000Z  200          20000
AAPL    2023-12-27T00:00:00.000000Z  240          24000

HAVING Clause

Filter grouped results:

SELECT 
    symbol,
    avg(price) AS avg_price,
    count() AS trade_count
FROM trades
GROUP BY symbol
HAVING count() > 100 AND avg(price) > 100;

Query Result:

symbol  avg_price  trade_count
AAPL    150.75     250
MSFT    250.25     200

Time-Series Aggregations

SAMPLE BY Aggregation

Aggregate by time intervals:

SELECT 
    timestamp,
    symbol,
    first(price) AS open,
    max(price) AS high,
    min(price) AS low,
    last(price) AS close,
    sum(quantity) AS volume
FROM trades
SAMPLE BY 1h;

Query Result:

timestamp                      symbol  open    high    low     close   volume
2023-12-28T10:00:00.000000Z    AAPL    150.00  151.00  149.50  150.75  15000
2023-12-28T11:00:00.000000Z    AAPL    150.80  152.00  150.50  151.50  12000
2023-12-28T10:00:00.000000Z    GOOG    140.00  142.00  139.50  141.50  8000

Rolling Aggregations

Combine with WINDOW JOIN for rolling calculations:

SELECT 
    t.symbol,
    t.timestamp,
    avg(p.price) AS rolling_avg_5min,
    stddev_samp(p.price) AS rolling_volatility
FROM trades t
WINDOW JOIN prices p ON (t.symbol = p.symbol)
RANGE BETWEEN 5 MINUTE PRECEDING AND CURRENT ROW;

Advanced Aggregation Patterns

Conditional Aggregation

Use CASE for conditional aggregation:

SELECT 
    symbol,
    sum(CASE WHEN price > 150 THEN quantity ELSE 0 END) AS high_price_volume,
    sum(CASE WHEN price <= 150 THEN quantity ELSE 0 END) AS low_price_volume
FROM trades
GROUP BY symbol;

Query Result:

symbol  high_price_volume  low_price_volume
AAPL    15000              10000
GOOG    0                  18000

Multiple Aggregation Levels

Nested aggregations:

SELECT 
    symbol,
    avg(hourly_volume) AS avg_hourly_volume,
    max(hourly_volume) AS peak_hourly_volume
FROM (
    SELECT 
        symbol,
        timestamp,
        sum(quantity) AS hourly_volume
    FROM trades
    SAMPLE BY 1h
)
GROUP BY symbol;

Pivot-Style Aggregation

Create columns from row values:

SELECT 
    date_trunc('day', timestamp) AS day,
    sum(CASE WHEN symbol = 'AAPL' THEN quantity ELSE 0 END) AS aapl_volume,
    sum(CASE WHEN symbol = 'GOOG' THEN quantity ELSE 0 END) AS goog_volume,
    sum(CASE WHEN symbol = 'MSFT' THEN quantity ELSE 0 END) AS msft_volume
FROM trades
GROUP BY day;

Query Result:

day                          aapl_volume  goog_volume  msft_volume
2023-12-28T00:00:00.000000Z  25000        18000        20000
2023-12-27T00:00:00.000000Z  24000        17000        19000

Performance Tips

Vectorized Aggregation

QuestDB automatically uses SIMD vectorization for aggregates:

-- Automatically vectorized
SELECT symbol, avg(price), sum(quantity)
FROM trades
GROUP BY symbol;

Indexed GROUP BY

Use SYMBOL columns for faster GROUP BY:

CREATE TABLE trades (
    symbol SYMBOL INDEX,  -- Indexed for fast grouping
    price DOUBLE,
    quantity LONG,
    timestamp TIMESTAMP
) TIMESTAMP(timestamp) PARTITION BY DAY;

Parallel Aggregation

QuestDB can parallelize aggregations across CPU cores automatically.

Filter Before Aggregation

Reduce data volume before aggregating:

SELECT symbol, avg(price)
FROM trades
WHERE timestamp > dateadd('d', -1, now())  -- Filter first
GROUP BY symbol;

NULL Handling

Aggregate functions handle NULL values:

count() - counts all rows
count(column) - counts non-NULL values
sum(), avg(), min(), max() - ignore NULL values
first(), last() - may return NULL
first_not_null(), last_not_null() - skip NULL values

SELECT 
    symbol,
    count() AS total_rows,
    count(price) AS non_null_prices,
    avg(price) AS avg_price
FROM trades
GROUP BY symbol;

Examples

Daily Trading Summary

SELECT 
    date_trunc('day', timestamp) AS day,
    symbol,
    first(price) AS open,
    max(price) AS high,
    min(price) AS low,
    last(price) AS close,
    sum(quantity) AS volume,
    count() AS trade_count
FROM trades
WHERE timestamp > dateadd('d', -7, now())
GROUP BY day, symbol
ORDER BY day DESC, symbol;

Market Statistics

SELECT 
    symbol,
    count() AS trades,
    sum(quantity) AS total_volume,
    avg(price) AS avg_price,
    stddev_samp(price) AS volatility,
    min(price) AS day_low,
    max(price) AS day_high,
    (max(price) - min(price)) / min(price) * 100 AS price_range_pct
FROM trades
WHERE timestamp > dateadd('d', -1, now())
GROUP BY symbol
ORDER BY total_volume DESC;

Hourly Volume Profile

SELECT 
    hour(timestamp) AS hour_of_day,
    avg(volume) AS avg_volume,
    max(volume) AS peak_volume
FROM (
    SELECT 
        timestamp,
        sum(quantity) AS volume
    FROM trades
    SAMPLE BY 1h
)
GROUP BY hour_of_day
ORDER BY hour_of_day;

Next Steps

Window Functions - Use analytical window functions
Time-Series Functions - Learn SAMPLE BY and LATEST ON
JOIN Operations - Combine data from multiple tables

Get Started

Core Concepts

Ingestion

SQL Guide

Operations

Deployment

Integrations

​Basic Aggregates

​COUNT

​SUM

​AVG

​MIN / MAX

​Statistical Aggregates

​STDDEV / STDDEV_SAMP

​STDDEV_POP

​VAR_SAMP

​VAR_POP

​COVAR_SAMP

​COVAR_POP

​CORR

​Positional Aggregates

​FIRST

​LAST

​FIRST_NOT_NULL

​LAST_NOT_NULL

​String Aggregates

​STRING_AGG

​Percentile Aggregates

​PERCENTILE_APPROX

​GROUP BY

​Single Column

​Multiple Columns

​HAVING Clause

​Time-Series Aggregations

​SAMPLE BY Aggregation

​Rolling Aggregations

​Advanced Aggregation Patterns

​Conditional Aggregation

​Multiple Aggregation Levels

​Pivot-Style Aggregation

​Performance Tips

​Vectorized Aggregation

​Indexed GROUP BY

​Parallel Aggregation

​Filter Before Aggregation

​NULL Handling

​Examples

​Daily Trading Summary

​Market Statistics

​Hourly Volume Profile

​Next Steps

Build docs developers (and LLMs) love

Basic Aggregates

COUNT

SUM

AVG

MIN / MAX

Statistical Aggregates

STDDEV / STDDEV_SAMP

STDDEV_POP

VAR_SAMP

VAR_POP

COVAR_SAMP

COVAR_POP

CORR

Positional Aggregates

FIRST

LAST

FIRST_NOT_NULL

LAST_NOT_NULL

String Aggregates

STRING_AGG

Percentile Aggregates

PERCENTILE_APPROX

GROUP BY

Single Column

Multiple Columns

HAVING Clause

Time-Series Aggregations

SAMPLE BY Aggregation

Rolling Aggregations

Advanced Aggregation Patterns

Conditional Aggregation

Multiple Aggregation Levels

Pivot-Style Aggregation

Performance Tips

Vectorized Aggregation

Indexed GROUP BY

Parallel Aggregation

Filter Before Aggregation

NULL Handling

Examples

Daily Trading Summary

Market Statistics

Hourly Volume Profile

Next Steps