nanoBars, nanoCubes, and nanoSlices are derived Nanotick | CME group historic data artefacts specifically designed for advanced high precision simulation and backtesting. nanoBars are available for all Comex, NYMEX, CBOT and CME futures and options contracts. The nanoBars historical data includes high precision, high frequency enriched bar data, full best bid and offer price history, all trade history and full limit order book history.
nanoBars & nanoCubes
The nanoBars data artefact is a hybrid of industry standard open-high-low-close-volume bar data enriched with additional fields and derived data that predicate upon Nanotick’s high precision timestamping technology. The fields in addition to standard OHLCV bars include high and low best bid and offers, buy and sell trade flow quantities, trade counts and VWAP along with high frequency book and trade flow alphas.
In addition to the data enrichment within the bar every nanoBar has an associated nanoCube. The nanoCube contains price and quantity for the top five book levels in full and then summary data for levels six to ten at the end of the bar time. This is then further extended by a series of deltas to the book image at standard offset intervals to the end of the bar.
The entire book depth as observed at the end of the bar time stamp is contained within the nanoSlice.
Unparalleled Simulation Accuracy
Both trading simulation and backtesting systems consuming nanoBars benefit from the additional data within nanoBars and their associated files to model both passive and aggressive fills to a level of precision that was previously only available to systems processing the entire tick stream in replay.
Simulating Aggressive Liquidity Removal
The nanoCube provides the requisite data for accurate determination of execution slippage as it contains the state of the book at different time shifts after the bar has expired. This enables both known and possible transmission latencies of the client system to be modelled. Switching between different latency offsets within the cube is trivial and enables profiling of key trading strategy metrics versus latency sensitivity. The nanoCube helps to quantify the impact of order transmission latency changes and “what if” scenarios about potential resource allocation, trading system latency optimisation and infrastructure enhancements with associated alpha and cost justifications. This is available without additional time and cost and the burden of complexity in processing raw tick streams from the exchange.
The nanoCube contains absolute and summary liquidity information down to ten levels for each time offset book. Fill rates and slippage can be better modelled for quantity as well as price. When aggressive orders penetrate deep into the book the average execution price and exact fill rate can be determined, thus enabling complex simulations to be performed such as when trading strategies are allowed to push close to the maximum achievable market share for a given market’s available liquidity. No other high performance processed data format allows for this level of accuracy and determinism.
Simulating Passive Liquidity Provision
nanoBars contain more than the industry standard high and low traded price for the width of the bar, they also include high and low best bid and offers. When simulating passive fills this additional data has a significant impact on strategy behaviour and helps avoid the trap of over-optimistic fills when testing.
For instance, when a market pulls back from support or resistance levels it is not uncommon for a small quantity “to print” at a traded price on the support or resistance level. It is frequently the case for there to be a significant quantity in the book at these support and resistance levels and the probability of our own simulated orders being at the front of the queue is certainly less than one.
When simulating passive fills with only high and low traded prices then the simulation risks filling passive orders at these price levels when in reality no fill would have been achieved. A common approach to simulation is to factor in the probability of a fill with some model of the spread and build in a “safety factor” whereby limit orders require a high or low price to be better than the order limit by some margin. Traditional bar data is devoid of spread information and this approach is at best a “poor guess” during fast markets when spreads widen and conversely opportunities may be at their best. By leveraging the extra fields within nanoBars simulation software can both accurately model the spread and also more pessimistically model passive fills by requiring the far touch price to take out resting orders, an operation which naturally factors in the spread at that time of day.
nanoBars are also delivered with a snapshot of the full current order book at the end of the bar. The nanoCube contains top of book information with time offset data whereas the nanoSlice is a complete snapshot containing price and quantity information for every price level populated in the market. Unlike the nanoCube there is no latency offset (time slippage) data with the nanoSlice but it does enable strategy developers to interrogate the entire liquidity in the market for every nanoBar both for alpha and capacity calculations.
nanoBars are available for all futures and options contracts for which Nanotick data is available and can be delivered in two artefact formats. Both formats contain the same information but each has different attributes that will impact data size, speed of replay and complexity of the API used to read the artefact.
This is the easiest artefact type to start with. nanoBars are delivered in a csv file where each row represents a single nanoBar for a given security.
This artefact type is not human readable. The advantage of this approach is speed. It is very quick to read the data. There is no text to number conversion required like in the ASCII formats.
nanoBars are available in four different data packs to suit differing simulation requirements and strategy investment horizons. These are as follows...
High Frequency Packs
The HFT packs contain nanoCube data with latency offsets from the end of the bar with 0, 1, 5, 10 and 25 milliseconds. These are designed for higher frequency strategies most likely trading from colocation.
HFT1 Pack: contains nanoBars with widths/durations of 1, 5 and 10 seconds
HFT2 Pack: contains nanoBars with widths/durations of 10, 30 and 60 seconds
Low Frequency Packs
The low frequency packs contain nanoCube information with longer latency offsets from the end of the bar. These are at 0, 50, 100, 250 and 500 milliseconds. These are designed for longer term strategies potentially trading via an intermediary or with lengthy communication times.
LFT1 Pack: contains nanoBars with widths/durations of 1, 5 and 10 minutes
LFT2 Pack: contains nanoBars with widths/durations of 15, 30 and 60 minutes
BBO and Trades
Every nanoBar distribution is accompanied by a complete and continuous record of best bid and offer data ( BBO) and every trade including the exchange's own native aggressor trade side.
Every nanoBar distribution contains...
a. The nanoBar pack
b. The selected nanoCube data
c. The nanoSlice file
nanoBars are priced by the pack
We offer discounts for clients who want to take a bundle of both HFT or LFT packs or the full strip of all available data.
nanoBar packs are available per security group or by product complex for Commodities, Energies, Equities, FX, Metals and Treasuries, for futures or options